G-Scout Enterprise and Cloud Security at Etsy

Posted by on November 18, 2019

As companies are moving to the cloud, they are finding a need for security tooling to audit and analyze their cloud environments. Over the last few years, various tools have been developed for this purpose. We’ll look at some of them and consider the uses for them. Specifically, we’ll take a close look at G-Scout, a tool I developed while working at NCC Group to look for security misconfigurations in Google Cloud Platform (GCP); and G-Scout Enterprise, a new tool with the same purpose, but tailored to the needs of security engineers at Etsy. We’ll also consider G-Scout Enterprise’s role within an ecosystem of other cloud logging and monitoring tools used at Etsy.

Cloud environments have a convenient feature which you won’t get from on premise servers: they have APIs. It’s similar for all the major cloud providers. They have a REST API which provides information on what services are being used, what resources exist, and how they are configured. An authorized user can call these APIs through a command line tool, or programmatically through a client library.

Those APIs provide information which is useful for security purposes. A classic example is a storage bucket (S3, GCS, etc.) which has been made public. It could be publicly readable, or publicly writable. Since we can use the API to see the permissions on any bucket we own, we can look for misconfigured permissions. So we go through all the API data we have for all our storage buckets, and look for permissions assigned to allUsers, or allAuthenticatedUsers.

Here are some other common examples:

Configuration Scanning Tools

Rather than making API calls and processing the data ad hoc, you can create a framework. A tool that will allow you, with a single command, to run various API calls to gather data on diverse resources, and then programmatically look for misconfigurations in that data. And in the end, you can have the tool place the results into a human-readable HTML report which you can browse according to your whims.

Scout 2 does all of the above for Amazon Web Services (AWS). G-Scout was created with a similar solution in mind as Scout 2, but for GCP. After Scout 2 there have followed plenty of other examples. Some, like G-Scout, have been open source, and others are available for purchase.

These tools continue to evolve. It is becoming increasingly common for companies to use more than one cloud provider. With this trend we’ve seen the creation of multi-cloud tools. Scout Suite has replaced Scout 2. Inspec supports AWS, Azure, and GCP.

And some of them have added features. Forseti Inventory stores the data collected in a SQL database (I’ve moved G-Scout in a similar direction, as we’ll see later). Forseti Enforcer will actually make changes to match policies. 

These features are useful, but not so much to a consultant, since a consultant shouldn’t want any permissions aside from viewer permissions. Scout 2 was designed for consulting. The user can get viewer permissions, run the tool, and leave no trace. Forseti, on the other hand, requires Organization Admin permissions, and creates a database and other resources within the organization that is being audited.

Difficulties With G-Scout

But the same basic functionality remains at the core of each of these tools. When it came to G-Scout, that core functionality worked well for smaller companies, or those less committed to GCP. But when there are hundreds of projects, thousands of instances, and many other resources, it becomes difficult to go through the results. 

Adding to this difficulty is the presence of false positives. Any automated tool is going to turn up false positives. Context may make something that seems like a finding at first glance, instead turn out to be acceptable. To return to our public storage bucket example, there are some cases where the content in the bucket is intended to be public. You can even serve a simple HTML website from a storage bucket. So it tends to fall to a human to go through and figure out which are false positives. Since it takes time to fix real findings, and the false positives don’t go away, running the tool frequently to see what’s new becomes untenable.

Finally, at Etsy, many of the findings G-Scout would turn up had already been found by other means, which we will explore a bit below.

We have a tool called Reactor. There is a stackdriver log sink for the organization, and those logs (with filters applied) go to a PubSub topic. There’s a cloud function that subscribes to that topic, and when it finds logs that match any of a further set of filters (the alerting rules) then it triggers an alert.

So for example, if someone makes a storage bucket public, an alert will trigger as soon as the corresponding stackdriver log is generated, rather than waiting for someone to run G-Scout at some point.

Here’s a partial example of a stackdriver log. As an API call to check IAM permissions would, it has all the information we need to trigger an alert. We see the user that granted the permission (in this case a service account). And below the fold we would see which role was assigned and which user it was assigned to.

Another point where we are alerting on misconfigurations is resource creation. We use Terraform for infrastructure as code. Before a Terraform apply is run, we have a series of unit tests that will be run by the pipeline. The unit tester runs tests for many of the same events which we alert on with the stackdriver logs. This includes the common example of a bucket being made public.

This is another process that is not so useful for a security consultant. But it’s better to catch misconfigurations in this way, than in the way Scout 2 or G-Scout would catch them, since this will prevent them from ever being implemented!

So we have what I’ll call a three-pronged approach to catching misconfigurations in GCP. These are the three prongs:

In summary, G-Scout’s traditional purpose was proving minimally useful. It was difficult to make good use of the G-Scout reports. And as we’ve seen, the first two prongs will usually catch misconfigurations first. So I moved away from G-Scout, and toward a new creation: G-Scout Enterprise.

G-Scout Enterprise

The fundamental change is to replace the HTML report with a BigQuery data collection. In fact, at its core, G-Scout Enterprise is very simple. It’s mostly just something that takes API data and puts it into BigQuery. Then other systems can do with that data as they please. The rules that will trigger alerts can be written in our alerting system like any other alerts we have (though they can also easily be written in Python within G-Scout Enterprise). We are now putting all of our other data into BigQuery as well, so it’s all connected.

Users can query any of the tables, each of which corresponds to one GCP API endpoint. G-Scout Enterprise tables can be joined – and they can be joined to our other data sources as well. And we can be very specific: like looking for all roles where amellos@etsy.com is a member, without enshrining it in our ruleset, because we can run queries through the BigQuery console. Or we can run queries in the command line, with helper functions that allow us to query with Python rather than SQL.

We can make comparisons and track changes over time. It can also provide context to other alerts. For example, if we have an IP address from an SSH alert, we can get information about the instance which owns that IP address, such as what service account it has, or what Chef role it has. 

Or for instance, the following, more complicated scenario:

We run Nessus. Nessus is an automated vulnerability scanner. It has a library of vulnerabilities it looks for by making network requests. You give it a list of IPs and it goes through them all. We now have it running daily. With a network of any size the volume of findings will quickly become overwhelming. Many of them are informational or safely ignored. But the rest need to be triaged, and addressed in a systematic way.

Not all Nessus findings are created equal. The same vulnerability on two different instances may be much more concerning on one than the other: if one is exposed to the internet and the other is not; if one is working with PII and the other is not; if one is in development and the other in production, and so on. Most of the information which determines how concerned we are with a vulnerability can be found among the collection of data contained in G-Scout Enterprise. This has simplified our scanning workflow. Since we can do network analysis with the data in G-Scout Enterprise, we can identify which instances are accessible from where. That means we don’t have to scan from different perspectives. And it has improved the precision of our vulnerability triaging, since there is so much contextual data available.

So we go through the following process:

  1. Enumerate all instances in our GCP account.
  2. Discard duplicate instances (instances from the same template, e.g. our many identical web server instances).
  3. Run the Nessus scan and place the results into BigQuery.
  4. Create a joined table of firewall rules and instances which they apply to (matching tags).
  5. Take various network ranges (, our corporate range, etc.), and for each firewall rule see if it allows traffic from that source.
  6. For instances with firewall rules that allow ingress from, see if the instance has a NatIP or is behind an external load balancer.
  7. Check whether the project the instance lives in is one of the projects classified as sensitive.
  8. Compute and assign scores according to the previous steps

And then we save the results into BigQuery. That gives us historical data. We can see if we are getting better or worse. We can see if we have certain troublemaker projects. We can empower our patch management strategy with a wealth of data.


That leaves us with a few main lessons gained from adapting G-Scout to Etsy:

One last note is that we have plans to open source G-Scout Enterprise in the coming months.

No Comments

Engineering Career Development at Etsy

Posted by on October 2, 2019 / No Responses

In late May of 2018, Etsy internally released an Engineering Career Ladder. Today, we’re sharing that ladder publicly and detailing why we decided to build it, why the content is what it is, and how it’s been put into use since its release.

Take a look

Defining a Career Ladder

A career ladder is a tool to help outline an engineer’s path for growth within a company. It should provide guidance to engineers on how to best take on new responsibilities, and allow their managers to assess and monitor performance and behavior. A successful career ladder should align career progression with a company’s culture, business goals, and guiding principles and act as a resource to guide recruiting, training, and performance assessments.

Etsy has had several forms of a career ladder before this iteration. The prior career ladders applied to all Etsy employees, and had a set of expectations for every employee in the same level across all disciplines. Overall, these previous ladders worked well for Etsy as a smaller company, but as the engineering team continued to grow we found the content needed updating to meet practical expectations, as the content in the ladder started to feel too broad and unactionable.

As a result, we developed this career ladder, specific to engineering, to allow us to be more explicit with those expectations and create a unified understanding of what it means to be an engineer at a certain level at Etsy. This ladder has been in place for over a year now, and in that time we’ve gone through performance reviews, promotion cycles, lots of hiring, and one-on-one career development conversations. We’re confident that we’ve made a meaningful improvement to engineering career development at Etsy and hope that releasing this career ladder publicly can help other companies support engineering career growth as well.

Designing the Etsy Engineering Career Ladder

We formed a working group focused on creating a new iteration of the career ladder comprised of engineers and engineering managers of various levels. The working group included Miriam Lauter, Dan Auerbach, and Jason Wain, and me. We started by exploring our current company-wide career ladder, discussing its merits and limitations, and the impact it had on engineering career development. We knew that any new version needed to be unique to Etsy, but we spent time exploring publicly available ladders of companies who had gone through a similar process in an effort to understand both tactical approaches and possible formats. Many thanks specifically to Spotify, Kickstarter, Riot Games, and Rent the Runway for providing insight into their processes and outcomes. Reviewing their materials was invaluable.

We decided our first step was to get on the same page as to what our goals were, and went through a few exercises resulting in a set of tenets that we felt would drive our drafting process and provide a meaningful way to evaluate the efficacy of the content. These tenets provided the foundation to our approach for developing the ladder.

The Tenets

Support meaningful career growth for engineers

Our career ladder should be clear enough, and flexible enough, to provide direction for any engineer at the company. We intended this document to provide actionable steps to advance your career in a way that is demonstrably impactful. Ideally, engineers would use this ladder to reflect on their time at Etsy and say “I’ve developed skills here I’ll use my entire career.”

Unify expectations across engineering

We needed to build alignment across the entire engineering department about what was required to meet the expectations of a specific level. If our career ladder were too open to interpretation it would cause confusion, particularly as it relates to the promotion process. We wanted to ensure that everyone had a succinct, memorable way to describe our levels, and understand exactly how promotions happen and what is expected of themselves and their peers.

Recognize a variety of valid career paths

Whether you’re building machine learning models or localizing our products, engineering requires skills across a range of competencies, and every team and project takes individuals with strengths in each. We wanted to be explicit about what we believe about the discipline, that valid and meaningful career paths exist at all levels for engineers who bring differences of perspectives and capabilities, and that not everyone progresses as an engineer in the same way. We intended to codify that we value growth across a range of competencies, and that we don’t expect every person to have the same set of strengths at specific points in their career.

Limit room for bias in how we recognize success

A career ladder is one in a set of tools that can help an organization mitigate potential bias. We needed to be thoughtful about our language, ensuring that it is inclusive, objective, and action oriented. We knew the career ladder would be used as basis for key career advancement moments, such as hiring and promotions, so developing a clear and consistent ladder was critical for mitigating potential bias in these processes.

Developing the Etsy Engineering Career Ladder

With these tenets in place, we had the first step towards knowing what was necessary for success. In addition to creating draft ladder formats, we set about determining how we could quantify the improvements that we were making. We outlined key areas where we’d need to directly involve our stakeholders, including engineering leadership, HR, Employee Resource Groups, and of course engineers. We made sure to define multiple perspectives for which the ladder should be a utility; e.g. an engineer looking to get promoted, a manager looking to help guide an engineer to promotion, or a manager who needed to give constructive performance feedback.

Implicit biases can be notoriously difficult to acknowledge and remove from these processes, and we knew that in order to do this as best as possible we’d need to directly incorporate feedback from many individuals, both internal and external, across domains and disciplines, and with a range of perspectives, to assure that we were building those perspectives into the ladder.

Our tactics for measuring our progress included fielding surveys and requests for open feedback, as well as direct 1:1 in-depth feedback sessions and third party audits to ensure our language was growth-oriented and non-idiomatic. We got feedback on structure and organization of content, comprehension of the details within the ladder, the ladder’s utility when it came to guiding career discussions, and alignment with our tenets.

The feedback received was critical in shaping the ladder. It helped us remove duplicative, unnecessary, or confusing content and create a format that we thought best aligned with our stated tenets and conveyed our intent. 

And finally, the Etsy Engineering Career Ladder

You can find our final version of the Etsy Engineering Career Ladder here.

The Etsy Engineering Career Ladder is split into two parts: level progression and competency matrix. This structure explicitly allows us to convey how Etsy supports a variety of career paths while maintaining an engineering-wide definition of each level. The level progression is the foundation of the career ladder. For each level, the ladder lays out all requirements including expectations, track record, and competency guidelines. The competency matrix lays out the behaviors and skills that are essential to meeting the goals of one’s role, function, or organization.

Level Progression

Each section within the level progression provides a succinct definition of the requirements for an engineer with that title. It details a number of factors, including the types of problems an engineer is solving, the impact of their work on organizational goals and priorities and how they influence others that they work with. For levels beyond Engineer I, we outline an expected track record, detailing achievements over a period of time in both scale and complexity. And to set expectations for growth of competencies, we broadly outline what levels of mastery an engineer needs to achieve in order to be successful.


If the level progression details what is required of an engineer at a certain level, competencies detail how we expect they can meet those expectations. We’ve outlined five core competency areas:

For each of these five competency areas, the competency matrix provides a list of examples that illustrate what it means to have achieved various levels of mastery. Mastery of a competency is cumulative — someone who is “advanced” in problem solving is expected to retain the skills and characteristics required for an “intermediate” or “beginner” problem solver.

Evaluating our Success

We internally released this new ladder in May of 2018. We did not immediately make any changes to our performance review processes, as it was critical to not change how we were evaluating success in the middle of a cycle. We merely released it as a reference for engineers and their managers to utilize when discussing career development going forward. When our next performance cycle kicked off, we began incorporating details from the ladder into our documentation and communications, making sure that we were using it to set the standards for evaluation.

Today, this career ladder is one of the primary tools we use for guiding engineer career growth at Etsy. Utilizing data from company-wide surveys, we’ve seen meaningful improvement in how engineers see their career opportunities as well as growing capabilities for managers to guide that growth.

Reflecting on the tenets outlined at the beginning of the process allows us to look back at the past year and a half and recognize the change that has occurred for engineers at Etsy and evaluate the ladder against the goals we believed would make it a success. Let’s look back through each tenet and see how we accomplished it.

Support meaningful career growth for engineers

While the content is guided by our culture and Guiding Principles, generally none of the competencies are Etsy-specific. The expectations, track record, and path from “beginner” to “leading expert” in a competency category are designed to show the growth of an engineer’s impact and recognize accomplishments that they can carry throughout their career, agnostic of their role, team, or even company.

The competency matrix also allows us to guide engineer career development within a level. While a promotion to a new level is a key milestone that requires demonstration of meeting expectations over time, advancing your level of mastery by focusing on a few key competencies allows engineers to demonstrate continual growth, even within the same level. This encourages engineers and their managers to escape the often insurmountable task of developing a plan to achieve the broader set of requirements for the next promotion, and instead create goals that help them get there incrementally.

Compared to our previous ladder, the path to Staff Engineer is no longer gated by the necessity to increase one’s breadth. We recognized that every domain has significantly complex, unscoped problems that need to be solved, and that we were limiting engineer growth by requiring those who were highly successful in their domain to expand beyond it. Having expectations outlined as they are now allows engineers the opportunity to grow by diving more deeply into their current domains.

Unify expectations across engineering

The definition for each level consists only of a few expectations, a track record, and guidelines for level of mastery of competencies. It is easy to parse, and to refer back to to get a quick understanding of the requirements. With a little reflection, it should be easy to describe how any engineer meets the three to five expectations of their level.

Prior to release, we got buy-in from every organizational leader in engineering that these definitions aligned with the reality of the expectations of engineers in their org. Since release we’ve aligned our promotion process to the content in the ladder. We require managers to outline how a candidate has met the expectations over the requisite period stated in the track record for their new level, and qualify examples of how they demonstrate the suggested level of mastery for competencies.

Recognize a variety of valid career paths

We ask managers to utilize the competencies document with their reports’ specific roles in mind when talking about career progression. Individual examples within the competency matrix may feel more or less applicable to individual roles, such as a Product Engineer or a Security Engineer, and this adaptability allows per-discipline growth while still aligning with the behaviors and outcomes we agree define a level of mastery. A small set of example skills is provided for each competency category that can help to better contextualize the application of the competencies in various domains. Additionally, we intentionally do not detail any competencies for which success is reliant on your team or organization.

Allowing managers to embrace the flexibility inherent in the competency matrix and its level of mastery system has allowed us to universally recognize engineer growth as it comes in various forms, building teams that embrace differences and value success in all its shapes. Managers can grow more diverse teams, for instance, by being able to recognize engineering leaders who are skilled domain experts, driving forward technical initiatives, and other engineering leaders who are skilled communicators, doing the glue work and keeping the team aligned on solving the right problems. We recognize that leadership takes many forms, and that is reflected in our competency matrix.

Limit room for bias in how we recognize success

The career ladder is only a piece of how we can mitigate potential bias as an organization. There are checks and balances built into other parts of Etsy’s human resources processes and career development programs, but since a career ladder plays such a key role in shaping the other processes, we approached this tenet very deliberately.

The competencies are not personality based, as we worked to remove anything that could be based on subjective perception of qualities or behaviors, such as “being friendly.” All content is non-idiomatic, in an effort to reduce differences in how individuals will absorb or comprehend the content. We also ensured that the language was consistent between levels by defining categories for each expectation. For instance, defining the expected complexity of the problems engineers solve per level allowed us to make sure we weren’t introducing any leaps in responsibility between levels that couldn’t be tied back to growth in the previous level. 

We also explicitly avoided any language that reads as quantifiable (e.g. “you’ve spoken at two or more conferences”) as opportunities to achieve a specific quantity of anything can be severely limited by your role, team, or personal situation, and can lead to career advice that doesn’t get at the real intent behind the competency. Additionally, evaluation of an individual against the ladder, for instance as part of a promotion, is not summarized in numbers. There is no score calculation or graphing an individual on a chart, nor is there an explicit number of years in role or projects completed as an expectation. While reducing subjectivity is key to mitigating potential bias, rigid numerical guidelines such as these can actually work against our other tenets by not allowing sufficient flexibility given an individual’s role.

Most importantly, the ladder was shaped directly through feedback from Etsy engineers, who have had direct personal experiences with how their individual situations may have helped or hindered their careers to draw on.

We’re really passionate about supporting ongoing engineer career growth at Etsy, and doing it in a way that truly supports our mission. We believe there’s a path to Principal Engineer for every intern and that this ladder goes a long way in making that path clear and actionable. We hope this ladder can serve as an example, in addition to those we took guidance from, to help guide the careers of engineers everywhere.

If you’re interested in growing your career with us, we’d love to talk, just click here to learn more.

No Comments

Apotheosis: A GCP Privilege Escalation Tool

Posted by on September 25, 2019

The Principle of Least Privilege

One of the most fundamental principles of information security is the principle of least privilege. This principle states that users should only be given the minimal permissions necessary to do their job. A corollary of the principle of least privilege is that users should only have those privileges while they are actively using them. For especially sensitive actions, users should be able to elevate their privileges within established policies, take sensitive actions, and then return their privilege level to normal to resume normal usage patterns. This is sometimes called privilege bracketing when applied to software, but it’s also useful for human users.

Following this principle reduces the chance of accidental destructive actions due to typos or misunderstandings. It may also provide some protection in case the user’s credentials are stolen, or if the user is tricked into running malicious code. Furthermore, it can be used as a notice to perform additional logging or monitoring of user actions.

In Unix this takes the form of the su command, which allows authorized users to elevate their privileges, take some sensitive actions, and then reduce their permissions. The sudo command is an even more fine-grained approach with the same purpose, as it will elevate privileges for a single command. 

Some cloud providers have features that allow for temporary escalation of privileges. Authorized users can take actions with a role other than the one which is normally assigned to them. The credentials used to assume a role are temporary, so they will expire after a specified amount of time. However, we did not find a built-in solution to achieve the same functionality in Google Cloud Platform (GCP).

Enter Apotheosis

Apotheosis is a tool that is meant to address the issues above. The word apotheosis means the elevation of someone to divine status. It’s possible, and convenient, to give users permanent “godlike” permissions, but this is a violation of the principle of least privilege. This tool will allow us to “apotheosize” users, and then return them to a “mortal” level of privilege when their job duties no longer require additional privileges.

Users or groups can be given “actual permissions” and “eligible permissions”. For example, a user who currently has the owner role may instead be given only the viewer role, and we will call that their “actual permissions”. Then we can give them “eligible permissions” of owner, which will come in the form of the service account token creator role on a service account with the editor or organization admin role.

For this user to elevate their privileges, the Apotheosis command line program will use their GCP credentials to call the REST API to create a short-lived service account token. Then, using that token, Apotheosis will make another REST API call which will grant the requested permissions to the user. Or, alternatively, the permissions may be granted to a specified third party, allowing the Apotheosis user to leverage their eligible permissions to grant actual permissions to another entity. The program will wait for a specified amount of time, remove the requested permissions, and then delete the short-lived service account token.

This process has the following advantages:

Future Additions

Some additional features which may be added to Apotheosis are contingent on the launch of other features, such as conditional IAM. Conditional IAM will allow the use of temporal restrictions on IAM grants, which will make Apotheosis more reliable. With conditional IAM, if Apotheosis is interrupted and does not revoke the granted permissions, they will expire anyway.

The ability to allow restricted permissions granting will be a useful IAM feature as well. Right now a user or service account can be given a role like editor or organization admin, and then can grant any other role in existence. But if it were possible to allow granting a predefined list of roles, that would make Apotheosis useful for a larger set of users. As it is now, Apotheosis is useful for users who have the highest level of eligible privilege, since their access to the Apotheosis service account gives them all the privileges of that service account. That is, the scope of those privileges can be limited to a particular project, folder, or organization, but cannot be restricted to a limited set of actions. At the moment that service account must have one of the few permissions which grant the ability to assign any role to any user. 

Requiring two-factor authentication when using the short-lived service account token feature on a particular service account would be another useful feature. This would require an Apotheosis user to re-authenticate with another factor when escalating privileges.

Open Source

Apotheosis is open source and can be found on Github.

No Comments

Code as Craft: Understand the role of Style in e-commerce shopping

Posted by and on August 2, 2019 / No Responses

Aesthetic style is key to many purchasing decisions. When considering an item for purchase, buyers need to be aligned not only with the functional aspects (e.g. description, category, ratings) of an item’s specification, but also its aesthetic aspects (e.g. modern, classical, retro) as well. Style is important at Etsy, where we have more than 60 million items and hundreds of thousands of them can differ by style and aesthetic. At Etsy, we strive to understand the style preferences of our buyers in order to surface content that best fits their tastes.

Our chosen approach to encode the aesthetic aspects of an item is to label the item with one of a discrete set of “styles” of which “rustic”, “farmhouse”, and “boho” are examples. As manually labeling millions of listings with a style class is not feasible – especially in a marketplace that is ever changing, we wanted to implement a machine learning model that best predicts and captures listings’ styles. Furthermore, in order to serve style-inspired listings to our users, we leveraged the style predictor to develop a mechanism to forecast user style preferences.

Style Model Implementation

Merchandising experts identified style categories.

For this task, the style labels are one of the classes that have been identified by our merchandising experts. Our style model is a machine learning model which, when given a listing and its features (text and images), can output a style label. The style model was designed to not only output these discrete style labels but also a multidimensional vector representing the general style aspects of a listing. Unlike a discrete label (“naval”, “art-deco”, “inspirational”) which can only be one class, the style vector encodes how a listing can be represented by all these style classes in varying proportions. While the discrete style labels can be used in predictive tasks to recommend items to users from particular style classes (say filtering recommended listings to a user from just “art-deco”), the style vector is supposed to serve as a machine learning signal into our other recommendation models. For example, on a listing page on Etsy, we recommend similar items. This model can now surface items that are not only functionally the same (“couch” for another “couch”) but can potentially recommend items that are instead from the same style (“mid-century couch” for a “mid-century dining table”).

The first step in building our listing style prediction model was preparing a training data set. For this, we worked with Etsy’s in-house merchandising experts to identify a list of 43 style classes. We further leveraged search visit logs to construct a “ground truth” dataset of items using these style classes. For example, listings that get a click, add to cart or purchase event for the search query “boho” are assigned the “boho” class label. This gave us a large enough labeled dataset to train a style predictor model.

Style Deep Neural Network

Once we had a ground truth dataset, our task was to build a listing style predictor model that could classify any listing into one of 43 styles (it is actually 42 styles and a ‘everything else’ catch all). For this task, we used a two layer neural network to combine the image and text features in a non-linear fashion. The image features are extracted from the primary image of a listing using a retrained Resnet model. The text features are the TF-IDF values computed on the titles and tags of the items. The image and text vectors are then concatenated and fed as input into the neural network model. This neural network model learns non-linear relationships between text and image features that best predict a listings style. This Neural Network was trained on a GPU machine on Google Cloud and we experimented with the architecture and different learning parameters until we got the best validation / test accuracy.

By explicitly taking style into account the nearest neighbors are more style aligned

User Style

As described above, the style model helps us extract low-dimension embedding vectors that capture this stylistic information for a listing, using the penultimate layer of the neural network. We computed the style embedding vector using the style model for all the listings in Etsy’s corpus.

Given these listing style embeddings, we wanted to understand users’ long-term style preferences and represent it as a weighted average of 42 articulated style labels. For every user, subject to their privacy preferences, we first gathered the entire history of “purchased”, “favorited”, “clicked” and “add to cart” listings in the past three months. From all these listings that a user interacted with, we combined their corresponding style vectors to come up with a final style representation for each user (by averaging them).

Building Style-aware User Recommendations

There are different recommendation modules on Etsy, some of which are personalized for each user. We wanted to leverage user style embeddings in order to provide more personalized recommendations to our users. For recommendation modules, we have a two-stage system: we first generate a candidate set, which is a probable set of listings that are most relevant to a user. Then, we apply a personalized ranker to obtain a final personalized list of recommendations.  Recommendations may be provided at varying levels of personalization to a user based on a number of factors, including their privacy settings.

In this very first iteration of user style aware recommendations, we apply user style understanding to generate a candidate set based on user style embeddings and their latest interacted taxonomies. This candidate set is used for Our Picks For You module on the homepage. The idea is to combine the understanding of a user’s long time style preference with his/her recent interests in certain taxonomies.

This work can be broken down into three steps:

Given user style embeddings, we take top 3 styles with the highest probability to be the “predicted user style”. Latest taxonomies are useful because they indicate users’ recent interests and shopping missions.

Given a taxonomy, sort all the listings in this taxonomy by the different style prediction scores for different classes, high to low. We take the top 100 listings out of these.

Minimal” listings in “Home & Living”

Floral” listings in “Home & Living”

Taxonomy, style validation is to check whether a style makes sense for a certain taxonomy. eg. Hygge is not a valid style for jewelry.

These become the style based recommendations for a user.

1-4: boho + bags_and_purses.backpacks
5-7: boho + weddings.clothing
8,13,16: minimal + bags_and_purses.backpacks

Style Analysis 

We were extremely interested to use our style model to answer questions about users sense of style. Our questions ranged from “How are style and taxonomy related? Do they have a lot in common?”, “Do users care about style while buying items?” to “How do style trends change across the year?”. Our style model enables us to answer at least some of these and helps us to better understand our users. In order to answer these questions and dig further we leveraged our style model and the generated embeddings to perform analysis of transaction data.

Next, we looked at the seasonality effect behind shopping of different styles on Etsy. We began by looking at unit sales and purchase rates of different styles across the year. We observed that most of our styles are definitely influenced by seasonality. For example, “Romantic” style peaks in February because of Valentines Day and “Inspirational” style peaks during graduation season. We tested the unit sales time series of different styles for statistical time series-stationarity test and found that the majority of the styles were non-stationary. This signifies that the majority of styles show different shopping trends throughout the year and don’t have constant unit sales throughout the year. This provided further evidence that users tastes show different trends across the year.

Using the style embeddings to study user purchase patterns not only provided us great evidence that users care about style, but also inspired us to further incorporate style into our machine learning products in the future.

Etsy is a marketplace for millions of unique and creative goods. Thus, our mission as machine learning practitioners is to build pathways that connect the curiosity of our buyers with the creativity of our sellers. Understanding both listing and user styles is another one of our novel building blocks to achieve this goal.

For further details into our work you can read our paper published in KDD 2019.

Authors: Aakash Sabharwal, Jingyuan (Julia) Zhou & Diane Hu

No Comments

An Introduction to Structured Data at Etsy

Posted by on July 31, 2019 / 1 Comment

Etsy has an uncontrolled inventory; unlike many marketplaces, we offer an unlimited array of one-of-a-kind items, rather than a defined set of uniform goods. Etsy sellers are free to list any policy-compliant item that falls within the three broad buckets of craft supplies, handmade, and vintage. Our lack of standardization, of course, is what makes Etsy special, but it also makes learning about our inventory challenging. That’s where structured data comes in.

Structured vs. Unstructured Data

Structured data is data that exists in a defined relationship to other data. The relation can be articulated through a tree, graph, hierarchy, or other standardized schema and vocabulary. Conversely, unstructured data does not exist within a standardized framework and has no formal relationship to other data in a given space.

For the purposes of structured data at Etsy, the data are the product listings, and they are structured according to our conception of where in the marketplace they belong. That understanding is expressed through the taxonomy.

Etsy’s taxonomy is a collection of hierarchies comprised of 6,000+ categories (ex. Boots), 400+ attributes (ex. Women’s shoe size), 3,500+ values (ex. 7.5), and 90+ scales (ex. US/Canada). These hierarchies form the foundation of 3,500+ filters and countless category-specific shopping experiences on the site. The taxonomy imposes a more controlled view of the uncontrolled inventory — one that engineers can use to help buyers find what they are looking for. 

Building the Taxonomy

The Etsy taxonomy is represented in JSON files, with each category’s JSON containing information about its place in the hierarchy and the attributes, values, and scales for items in that category. Together, these determine what questions will be asked of the seller for listings in that category (Figure A, Box 1), and what filters will be shown to buyers for searches in that category (Figure A, Box 2).

Figure A 
A snippet of the JSON representation of the Jewelry > Rings > Bands category

The taxonomists at Etsy are able to alter the taxonomy hierarchies using an internal tool. This tool supports some unique behaviors of our taxonomy, like inheritance. This means that if a category has a particular filter, then all of its subcategories will inherit that filter as well.

Figure B
Sections of the Jewelry > Rings > Bands category as it appears in our internal taxonomy tool 

Gathering Structured Data: The Seller Perspective

One of the primary ways that we currently collect structured data is through the listing creation process, since that is our best opportunity to learn about each listing from the person who is most familiar with it: the seller!

Sellers create new listings using the Shop Manager. The first step in the listing process is to choose a category for the listing from within the taxonomy. Using auto-complete suggestions, sellers can select the most appropriate category from all of the categories available. 

Figure C 
Category suggestions for “ring”

At this stage in the listing creation process, optional attribute fields appear in the Shop Manager. This is also enabled by the taxonomy JSON, in that the fields correspond with the category selected by the seller (see Figure A, Box 1). This behavior ensures that we are only collecting relevant attribute data for each category and simplifies the process for sellers. Promoting this use of standardized data also reduces the need for overloaded listing titles and descriptions by giving sellers a designated space to tell buyers about the details of their products. Data collected during the listing creation process appears on the listing page, highlighting for the buyer some of the key, standardized details of the listing.

Figure D
Some of the attribute fields that appear for listings in Jewelry > Rings > Bands (see Figure A, Box 1 for the JSON that powers the Occasion attribute)

Making Use of Structured Data: The Buyer Perspective

Much of the buyer experience is a product of the structured data that has been provided by our sellers. For instance, a given Etsy search yields category-specific filters on the left-hand navigation of the search results page. 

Figure E
Some of the filters that appear upon searching for “Rings”

Those filters should look familiar! (see Figure D) They are functions of the taxonomy. The search query gets classified to a taxonomy category through a big data job, and filters affiliated with that category are displayed to the user (see Figure F below). These filters allow the buyer to narrow down their search more easily and make sense of the listings displayed.

Figure F
The code that displays category-specific filters upon checking that the classified category has buyer filters defined in its JSON (see Figure A, Box 2 for a sample filter JSON)

Structuring Unstructured Data

There are countless ways of deriving structured data that go beyond seller input. First, there are ways of converting unstructured data that has already been provided, like listing titles or descriptions, into structured data. Also, we can use machine learning to learn about our listings and reduce our dependence on seller input. We can, for example, learn about the color of a listing through the image provided; we can also infer more nuanced data about a listing, like its seasonality or occasion.

We can continue to measure the relevance of our structured data through metrics like the depth of our inventory categorization within our taxonomy hierarchies and the completeness of our inventory’s attribution.

All of these efforts allow us to continue to build deeper category-specific shopping experiences powered by structured data. By investing in better understanding our inventory, we create deeper connections between our sellers and our buyers.

1 Comment

What it’s Like to Intern at Etsy? – Part I

Posted by on March 18, 2019 / 1 Comment

I secretly like seeing people’s surprise when I told them that I chose to intern at Etsy because it was the only company that asked for a cover letter. I enjoyed every second of filling out my Etsy software engineering internship application because I felt like I was really telling my story to a company that cared about my whole self. I interned at Etsy during summer 2016 and started working full-time after I graduated from college in 2017. The human touch embedded in Etsy’s engineering culture, business strategy and company vision is still the number one thing I am proud of.

Over the past three years, I have gotten many questions about what it’s like to intern and have my first job out of college at Etsy. It always gives me a warm feeling when students are curious and excited about careers at Etsy, and I think it’s time we give this question answers that will live on the interweb.

This past winter, I met five interns that Etsy hosted for WiTNY (Women in Technology and Entrepreneurship in New York)’s Winternship program. At the end of their three-week internships, they were super excited to share their experiences. One of the winterns, Nia Laureano, wrote a fantastic recap of her time at Etsy, and I thought it would be a great way to start sharing the Etsy internship experience!

Inventing a Process: Five Interns Navigate a Complex Problem Thanks to Etsy’s Relentlessly Human Touch

by Nia Laureano

Interning at Etsy is a unique experience because so much about Etsy’s identity has to be understood in order to move forward with any sort of work. For three weeks in January, a team of four girls and I joined the Etsy family as interns and were tasked with solving an issue being faced by a product team that helps the buyers.

Coming into our first day, the details of our task were overwhelmingly foreign to us. The subject we were dealing with was Etsy’s listing page. It’s a complicated page, due to the fact that 60 million listings exist on Etsy and they are all vastly different. When engineers make changes to the listing page, it is difficult to test their code against every possible variation of the page that exists. Sometimes, a variation slips from their mind — they forget to account for it, which could potentially cause the page to break when they push code. This is what engineers call an edge case, and our job was to create a tool that allows Etsy engineers to test for edge cases more thoroughly. Specifically, we were asked to create a reference for them to easily find listings that match different criteria and variations to test code against. But solving a project that we barely understood ourselves seemed daunting, if not impossible.

The entirety of our first week was spent immersing ourselves in the context of this world we were working in. We strolled through the typical workflow of an Etsy engineer, trying to imagine where our solution would fit neatly into the puzzle. We spoke to engineers about their frustrations to get to the root of their needs. We became engineers by being thrusted into the process of pushing code to Etsy’s repository. We couldn’t commit to our craft without first understanding how these employees live and work; then we had to imagine what we could do to make their world better.

After interviewing several engineers, we realized that they each had their own ways of testing for edge cases. “I just have two folders of bookmarks that have some links in them,” said one engineer. “But I’m not sure what other people use.” It was surprising to hear that engineers weren’t sure what other people on their team were doing. We realized, at this point, that the problem wasn’t a faulty process — there was no process to begin with. It was up to us to invent a process, or at least establish a basic standard when it comes to testing for edge cases.

In ideation, the solutions we envisioned ranged dramatically. Something as basic as a spreadsheet would have been helpful, but we also dreamed bigger. We thought about creating an automated Etsy shop that auto-generates listings that represent the edge cases that needed to be tested. We wanted to create something ambitious, but it also had to be something we could attain in three weeks. Ultimately, we focused on creating a solution that would deliver on three crucial needs of our engineers: structure, convenience and confidence.

Structure. While some engineers relied on their own bookmarks or spreadsheets to keep track of edge cases, some relied on sheer memory, or asking their coworkers via Slack. Testing for something that could potentially break the listing page, we realized, shouldn’t be such a structureless process. Our solution needed to provide an element of uniformity; it needed to eliminate that glaring unawareness about what other teammates were doing. It needed to be a unifier.

Convenience. In order to make a tool that was accessible and easy to use, we needed to identify and understand the environment in which engineers complete the bulk of their work, because that’s where we would want our tool to live. We quickly noticed one common thread woven through the workflow of not only Etsy’s engineers, but the company as a whole: our messaging platform, Slack. We observed that so much important work at Etsy is already accomplished via Slack; it’s where employees collaborate and even push code. It made perfect sense for our solution to be integrated within the environment that was already so lived-in.

Confidence. Bugs are inevitable, but our engineers deserve to feel confident that the code they are pushing is as clean as it can be. The more edge cases they can test for, the more certain they can feel that their code is quality and fully functional. Therefore, our solution had to be thorough and reliable; it had to be something engineers could trust.

After three weeks, our project was completed in two phases. Our first phase was focused on creating a spreadsheet. This was the skeleton of our final product, which mirrored the anatomy of the listing page itself. To build this, we broke down the different components of the listing page and identified all of the variations that could occur within those components. Then, we spent several days creating almost one hundred of our own listings on Etsy that represented each of those variations. We ended up with a thorough, intuitively structured catalog of edge cases which can now be accessed by anyone at Etsy who needs it.

The second phase of our project was a Slack-integrated bot. Using our spreadsheet as a backbone, we aimed to design a bot that can retrieve edge cases on command via Slack. Engineers can input commands that return single, multiple, or all edge cases they may be looking for. Due to our time constraint, we were only able to create a bot that utilizes test data, but we hope to see a future iteration that fully integrated with our spreadsheet.

A universe of terminology and culture had to be packed into our brains in order to accomplish what we did in three weeks. Yet, we somehow felt so seamlessly integrated into Etsy’s ecosystem from day one, thanks to the friendly and enthusiastic nature of everyone around us. We were never afraid to ask questions, because no one ever talked down to us or made us feel inferior. There are no mechanisms in place at Etsy that make power dynamics apparent, not even from the perspective of an intern.

Our project was completed not because of crash courses in PHP or because we overloaded on cold brew; it was thanks to the people who nurtured us along the way. It was the prospect of creating something that could make a lasting impact on a company we loved that motivated us. Etsy’s relentlessly human touch makes even the smallest of projects feel meaningful, and it can turn three weeks into an unforgettable experience that I will never stop feeling passionate about.

A note about our internship & our organization:

WiTNY (Women in Technology and Entrepreneurship in New York) is a collaborative initiative between Cornell Tech x CUNY designed to inspire young women to pursue careers in technology. WiTNY offers workshops and program that teach important skills and provide real work experience.

The Winternship program is a paid, three-week, mini-internship for first and second-year undergraduate students at CUNY schools, during their January academic recess. Etsy is one of many companies who participated in the Winternship program this year, taking a team of five young women and giving them a challenging project to complete while also teaching them about the different roles within a tech company.


1 Comment

Executing a Sunset

Posted by on February 1, 2019 / 5 Comments

We all know how exciting it is to build new products, the thrill of a pile of new ideas waiting to be tested, new customers to reach, knotty problems to solve, and dreams of upward-sloping graphs.  But what happens when it is no longer aligned with the trajectory of the company. Often, the product, code, and infrastructure become a lower priority, while the team moves on to the next exciting new venture. In 2018, Etsy sunset Etsy Wholesale, Etsy Studio, and Etsy Manufacturing, three customer-facing products.

In this blog post, we will explore how we sunset these products at Etsy. This process involves a host of stakeholders including marketing, product, customer support, finance and many other teams, but the focus of this blog post is on engineering and the actual execution of the sunset.

ProcessPre-code deletion

Use Feature Flags and Turn off Traffic

Once the communication had been done through emails, in-product announcements, and posts in the user forums, we started focusing on the execution. Prior to the day of each sunset, we used our feature flagging infrastructure to build a switch to disable access to the interface for Wholesale and Manufacturing. Feature flags are an integral part of the continuous deployment process at Etsy. Feature flags reinforce the benefits of small changes and continuous delivery.

On the day of the sunset, all we had to do was deploy a one line configuration change and the product was shut off since there was a feature flag that controlled access to these products.

A softer transition is often preferable to a hard turn off. For example, we disabled the ability for buyers to create new orders one month before shutting Etsy Wholesale off. That gave sellers a chance to service the orders that remained on-platform, avoiding a mad-dash at the end.

Export Data for Users

Once the Etsy Wholesale platform was turned off, we created data export files for each seller and buyer with information about every order they received or placed during the five years that the platform was active. Generating and storing these files in one shot allowed us to clean up the wholesale codebase without fear that parts of it would be needed later for exporting data.

Set Up Redirects

We highly recommend redirects through feature flags,  but a hard DNS redirect might be required in some circumstances. The sunset of Etsy Studio was complicated by the fact that in the middle of this project, etsy.com was being migrated from on-premise hosting to the cloud. To reduce complexity and risk for the massive cloud migration project, Etsy Studio had to be shut off before the migration began. On the day before the cloud migration, a DNS redirect was made to forward any request on etsystudio.com to a special page on etsy.com that explained that Etsy Studio was being shut down. Once the DNS change went live, it effectively shut off Etsy Studio completely.

Code Deletion Methodology:

Once we confirmed that all three products were no longer receiving traffic, we kicked off the toughest part of the engineering process: deleting all the code. We tried to phase it in two parts, as tightly and loosely integrated products. Integrations in the most sensitive/dangerous spots were prioritized, and safer deletions were done later as we were heading into the holiday season (our busiest time of the year).

For Etsy Wholesale and Etsy Manufacturing, we had to remove the code piece-by-piece because it was tightly integrated with other features on the site. For Etsy Studio, we thought we would be able to delete the code in one massive commit. One benefit of our continuous integration system is that we can try things out, fail, and revert without negatively affecting our users. This proved valuable as when we tried deleting the code one massive commit, some unit tests for the Etsy codebase started failing. We realized that small dependencies between the code had formed over time. We decided to delete the code in smaller, easier to test, chunks.

a small example of dependencies creeping in where you least expect them.

Challenges: Planning (or lack of it) for slowdowns


During the process of sunsetting, we didn’t consider how busy other teams would be heading into the holiday season. This slowed down the process of getting code reviews approved. This became especially crucial for us since we were removing and modifying big chunks of code maintained by other teams.

There were also several other big projects in flight while we were trying to delete code across our code base and that slowed us down. One example that I already mentioned was cloud migration: we couldn’t shut off Etsy Studio using a config flag and we had to work around it.

Commit Size and Deploys

To reduce risk, our intention was to keep our commits small, but when trying to delete so much code at once, it’s hard to keep all your commits small. Testing and deploying was a least 50% of our team’s time. Our team made about 413 commits over five months, deleting 275,000 lines of code. That averages out to 630 lines of code deleted per commit, which were frequently deployed one at a time.


We actively think of compliance when building new things, but it is also important to keep in mind compliance requirements when you delete code. Etsy’s SOX compliance system requires that certain files in our codebase are subject to extra controls. When we deploy changes to such files, we need additional reviews and signoffs. We had to do 44 SOX reviews since we did multiple small commits. Each review requires approvals by multiple people and this added on average a few hours to each bit of deletion we did.  Similarly, we considered user privacy and data protection in how to make retention decisions about sunsetted products, how to make data available for export, and how it impacts our terms and policies.

Deleting so much code can be a difficult process. We had to revert changes from production at least five times, which, for the most part was simple. One of these five reverts was complicated by a data corruption issue affecting a small population of sellers, which required several days of work to write, test, and run a script to fix the problem.

The Outcome

We measured success using the following metrics:

From 1000s of error logs a day for wholesale, to less than 100 (eventually we got this to zero)

The roots that three products had in our systems demonstrated the challenges in building and maintaining a standalone product alongside our core marketplace. The many branching pieces of logic that snuck in made it difficult to reuse lots of existing code. By deleting 275,000 lines of code, we were able to reduce tech debt and remove roadblocks for other engineers.   


Why Diversity Is Important to Etsy

Posted by on January 7, 2019 / 2 Comments

We recently published our company’s Guiding Principles. These are five common guideposts that apply to all organizations and departments within Etsy. We spent a great deal of time discussing, brainstorming, and editing these. By one estimate, over 30% of the company had some input at some phase of the process. This was a lot of effort by a lot of people but this was important work. These principles need to not only reflect how we currently act but at the same time they need to be aspirational for how we want to behave. These principles will be used in performance assessments, competency matrices, interview rubrics, career discussions, and in everyday meetings to refocus discussions.

One of the five principles is focused on diversity and inclusion. The principle states:

We embrace differences.

Diverse teams are stronger, and inclusive cultures are more resilient. When we seek out different perspectives, we make better decisions and build better products.

Why would we include diversity and inclusion as one of our top five guiding principles? One reason is that Etsy’s mission is to Keep Commerce Human. Etsy is a very mission-driven company. Many of our employees joined and remain with us because they feel so passionate about the mission. Every day, we keep commerce human by helping creative entrepreneurs find buyers who become committed fans of the seller’s art, crafts, and collections. The sellers themselves are a diverse group of individuals from almost every country in the world. We would have a hard time coming to work if the way we work, the way we develop products, the way we provide support, etc. isn’t done in a manner that supports this mission. Failing to be diverse and inclusive would fail that mission.

Besides aligning with our mission, there are other reasons that we want to have diverse teams. Complicated systems, which feature unpredictable, surprising, and unexpected behaviors have always existed. Complex systems, however, have gone from something found mainly in large systems, such as cities, to almost everything we interact with today. Complex systems are far more difficult to manage than merely complicated ones as subsystems interact in unexpected ways making it harder to predict what will happen. Our engineers deal with complex systems on a daily basis. Complexity is a bit of an overloaded term, but scholarly literature generally categorizes it into three major groups, determined according to the point of view of the observer: behavioral, structural, and constructive.1 Between the website, mobile apps, and systems that support development, our engineers interact with highly complex systems from all three perspectives every day. Research has consistently shown that diverse teams are better able to manage complex systems.2

We recently invited Chris Clearfield and András Tilcsik, the authors of Meltdown (Penguin Canada, 2018), to speak with our engineering teams. The book and their talk contained many interesting topics, most based on Charles Perrow’s book, Normal Accident Theory (Princeton University Press; revised ed. 1999). However, perhaps the most important topic was based on a series of studies performed by Evan Apfelbaum and his colleagues at MIT. This study revealed that as much as we’re predisposed to agree with a group, our willingness to disagree increases dramatically if the group is diverse.3 According to Clearfield and Tilcsik, homogeneity may facilitate “smooth, effortless interactions,” but diversity drives better decisions. Interestingly, it’s the diversity and not necessarily the specific contributions of the individuals themselves, that causes greater skepticism, more open and active dialogue, and less group-think. This healthy skepticism is incredibly useful in a myriad of situations. One such situation is during pre-mortems, where a project team imagines that a project has failed and works to identify what potentially could lead to such an outcome. This is very different from a postmortem where the failure has already occurred and the team is dissecting the failure. Often individuals who have been working on projects for weeks or more are biased with overconfidence and the planning fallacy. This exercise can help ameliorate these biases and especially when diverse team members participate. We firmly believe that when we seek out different perspectives, we make better decisions, build better products, and manage complex systems better.

Etsy Engineering is also incredibly innovative. One measure of that is the number of open source projects on our GitHub page and the continuing flow of contributions from our engineers in the open source community. We are of course big fans of open source as Etsy, like most modern platforms, wouldn’t exist in its current form without the myriad of people who have solved a problem and published their code under an open source license. But we also view this responsibility to give back as part of our culture. Part of everyone’s job at Etsy is making others better. It has at times been referred to as “generosity of spirit”, which to engineers means that we should be mentoring, teaching, contributing, speaking, writing, etc.  

Another measure of our innovation is our experiment velocity. We often run dozens of simultaneous experiments in order to improve the buyer and seller experiences. Under the mission of keeping commerce human, we strive every day to develop and improve products that enable 37M buyers to search and browse through 50M+ items to find just the right, special piece. As you can imagine, this takes some seriously advanced technologies to work effectively at this scale. And, to get that correct we need to experiment rapidly to see what works and what doesn’t. Fueling this innovation is the diversity of our workforce.

Companies with increased diversity unlock innovation by creating an environment where ideas are heard and employees can find senior-level sponsorship for compelling ideas. Leaders are twice as likely to unleash value-driving insights if they give diverse voices equal opportunity.4

So diversity fits our mission, helps manage complex systems, and drives greater innovation, but how is Etsy doing with respect to diversity? More than 50% of our Executive Team and half of our Board of Directors are women. More than 30% of Etsy Engineers identify as women/non-binary and more than 30% are people of color.5 These numbers are industry-leading, especially when compared to other tech companies who report “tech roles” and not the more narrow category, “engineering” roles. Even though we’re proud of our progress, we’re not fully satisfied. In October 2017, we announced a diversity impact goal to “meaningfully increase representation of underrepresented groups and ensure equity in Etsy’s workforce.” To advance our goal, we are focused on recruiting, hiring, retention, employee development, mentorship, sponsorship, and building an inclusive culture.

We have been working diligently on our recruiting and hiring processes. We’ve rewritten job descriptions, replaced some manual steps in the process with third-party vendors, and changed the order of steps in the interview process, all in an effort to recruit and hire the very best engineers without bias. We have also allocated funding and people in order to sponsor and attend conferences focused on underrepresented groups in tech. We’ll share our 2018 progress in Q1 2019.

Once engineers are onboard, we want them to bring their whole selves to work in an inclusive environment that allows them to thrive and be their best. One thing that we do to help with this is to promote and partner directly with employee resource groups (ERGs). Our ERGs include Asian Resource Community, Black Resource and Identity Group at Etsy, Jewish People at Etsy, Hispanic Latinx Network, Parents ERG, Queer@Etsy, and Women and NonBinary People in Tech. If you’re not familiar with ERGs, their mission and goals are to create a positive and inclusive workplace culture where employees from underrepresented backgrounds, lifestyles, and abilities have access to programs that foster a sense of community, contribute to professional development, and amplify diverse voices within our organization. Each of these ERGs has an executive sponsor. This ensures that there is a communication channel with upper management. It also highlights the value that we place upon the support that these groups provide.    

We are also focused on retaining our engineers. One of the things that we do to help in this area is to monitor for discrepancies that might indicate bias. During our compensation, assessment, and promotion cycles, we evaluate for inconsistencies. We perform this analysis both internally and through the use of third parties.  

Etsy Engineering has been a leader and innovator in the broader tech industry with regard to technology and process. We also want to be leaders in the industry with regards to diversity and inclusion. It is not only the right thing to do but it’s the right thing to do for our business. If this sounds exciting to you, we’d love to talk, just click here to learn more.



1 Wade, J., & Heydari, B. (2014). Complexity: Definition and reduction techniques. In Proceedings of the Poster Workshop at the 2014 Complex Systems Design & Management International Conference.
2 Sargut, G., & McGrath, R. G. (2011). Learning to live with complexity. Harvard Business Review, 89(9), 68–76
3 Apfelbaum EP, Phillips KW, Richeson JA (2014) Rethinking the baseline in diversity research: Should we be explaining the effects of homogeneity? Perspect Psychol Sci 9(3):235–244.
4 Hewlett, S. A., Marshall, M., & Sherbin, L. (2013). How diversity can drive innovation. Harvard Business Review.
5 Etsy Impact Update (August 2018). https://extfiles.etsy.com/Impact/2017EtsyImpactUpdate.pdf


boundary-layer : Declarative Airflow Workflows

Posted by on November 14, 2018

When Etsy decided last year to migrate our operations to Google Cloud Platform (GCP), one of our primary motivations was to enable our machine learning teams with scalable resources and the latest big-data and ML technologies. Early in the cloud migration process, we convened a cross-functional team between the Data Engineering and Machine Learning Infrastructure groups in order to design and build a new data platform focused on this goal.

One of the first choices our team faced was how to coordinate and schedule jobs across a menagerie of new technologies. Apache Airflow (incubating) was the obvious choice due to its existing integrations with GCP, its customizability, and its strong open-source community; however, we faced a number of open questions that had to be addressed in order to give us confidence in Airflow as a long-term solution.

First, Etsy had well over 100 existing Hadoop workflows, all written for the Apache Oozie scheduler. How would we migrate these to Airflow? Furthermore, how would we maintain equivalent copies of Oozie and Airflow workflows in parallel during the development and validation phases of the migration, without requiring our data scientists to pause their development work?

Second, writing workflows in Airflow (expressed in python as directed acyclic graphs, or DAGs) is non-trivial, requiring new and specialized knowledge. How would we train our dozens of internal data platform users to write Airflow DAGs? How would we provide automated testing capabilities to ensure that DAGs are valid before pushing them to our Airflow instances? How would we ensure that common best-practices are used by all team members? And how would we maintain and update those DAGs as new practices are adopted and new features made available?

Today we are pleased to introduce boundary-layer, the tool that we conceived and built to address these challenges, and that we have released to open-source to share with the Airflow community.

Introduction: Declarative Workflows

Boundary-layer is a tool that enables data scientists and engineers to write Airflow workflows in a declarative fashion, as YAML files rather than as python. Boundary-layer validates workflows by checking that all of the operators are properly parameterized, all of the parameters have the proper names and types, there are no cyclic dependencies, etc. It then translates the workflows into DAGs in python, for native consumption by Airflow.

Here is an example of a very simple boundary-layer workflow:

name: my-dag-1

  start_date: '2018-10-01'

- name: print-hello
  type: bash
    bash_command: "echo hello"
- name: print-world
  type: bash
  - print-hello
    bash_command: "echo world"

Boundary-layer translates this into python as a DAG with 2 nodes, each consisting of a BashOperator configured with the provided properties, as well as some auto-inserted parameters:

# Auto-generated by boundary-layer

import os
from airflow import DAG

import datetime

from airflow.operators.bash_operator import BashOperator

        'start_date': '2018-10-01',

dag = DAG(
        dag_id = 'my_dag_1',
        default_args = DEFAULT_TASK_ARGS,

print_hello = BashOperator(
        dag = (dag),
        bash_command = 'echo hello',
        start_date = (datetime.datetime(2018, 10, 1, 0, 0)),
        task_id = 'print_hello',

print_world = BashOperator(
        dag = (dag),
        bash_command = 'echo world',
        start_date = (datetime.datetime(2018, 10, 1, 0, 0)),
        task_id = 'print_world',


Note that boundary-layer inserted all of the boilerplate of python class imports and basic DAG and operator configuration. Additionally, it validated parameter names and types according to schemas, and applied type conversions when applicable (in this case, it converted date strings to datetime objects).


Moving from python-based to configuration-based workflows naturally imposes a functionality penalty. One particularly valuable feature of python-based DAGs is the ability to construct them dynamically: for example, nodes can be added and customized by iterating over a list of values. We make extensive use of this functionality ourselves, so it was important to build a mechanism into boundary-layer to enable it.

Boundary-layer generators are the mechanism we designed for dynamic workflow construction. Generators are complete, distinct sub-workflows that take a single, flexibly-typed parameter as input. Each generator must prescribe a mechanism for generating a list of values: for example, lists of items can be retrieved from an API via an HTTP GET request. The python code written by boundary-layer will iterate over the list of generator parameter values and create one instance of the generator sub-workflow for each value. Below is an example of a workflow that incorporates a generator:

name: my-dag-2

  start_date: '2018-10-01'

- name: retrieve-and-copy-items
  type: requests_json_generator
  target: sense-and-run
    url: http://my-url.com/my/file/list.json
    list_json_key: items

- name: print-message
  type: bash
  - retrieve-and-copy-items
    bash_command: echo "all done"
name: sense-and-run

- name: sensor
  type: gcs_object_sensor
    bucket: <<item['bucket']>>
    object: <<item['name']>>
- name: my-job
  type: dataproc_hadoop
    cluster_name: my-cluster
    region: us-central1
    main_class: com.etsy.jobs.MyJob
    - <<item['name']>>

This workflow retrieves the content of the specified JSON file, extracts the items field from it, and then iterates over the objects in that list, creating one instance of all of the operators in the sense-and-run sub-graph per object.

Note the inclusion of several strings of the form  << ... >>.  These are boundary-layer verbatim strings, which allow us to insert inline snippets of python into the rendered DAG. The item value is the sub-workflow’s parameter, which is automatically supplied by boundary-layer to each instance of the sub-workflow.

Also note that generators can be used in dependency specifications, as indicated by the print-message operator’s upstream_dependencies block. Generators can even be set to depend on other generators, which boundary-layer will encode efficiently, without creating a combinatorially-exploding set of edges in the DAG.

Advanced features

Under the hood, boundary-layer represents its workflows using the powerful networkx library, and this enables a variety of features that require making computational modifications to the graph, adding usability enhancements that go well beyond the core functionality of Airflow itself.

A few of the simpler features that modify the graph include before and after sections of the workflow, which allow us to specify a set of operators that should always be run upstream or downstream of the primary list of operators. For example, one of our most common patterns in workflow construction is to put various sensors in the before block, so that it is not necessary to specify and maintain explicit upstream dependencies between the sensors and the primary operators. Boundary-layer automatically attaches these sensors and adds the necessary dependency rules to make sure that no primary operators execute until all of the sensors have completed.

Another feature of boundary-layer is the ability to prune nodes out of workflows, while maintaining all dependency relationships between the nodes that remain. This was especially useful during the migration of our Oozie workflows. It allowed us to isolate portions of those workflows for running in Airflow and gradually add more portions in stages, until the workflows were fully migrated, without ever having to create the portioned workflows as separate entities.

One of the most useful advanced features of boundary-layer is its treatment of managed resources. We make extensive use of ephemeral, workflow-scoped Dataproc clusters on the Etsy data platform. These clusters are created by Airflow, shared by various jobs that Airflow schedules, and then deleted by Airflow once those jobs are complete. Airflow itself provides no first-class support for managed resources, which can be tricky to configure properly: we must make sure that the resources are not created before they are needed, and that they are deleted as soon as they are not needed anymore, in order to avoid accruing costs for idle clusters. Boundary-layer handles this automatically, computing the appropriate places in the DAG into which to splice the resource-create and resource-destroy operations. This makes it simple to add new jobs or remove old ones, without having to worry about keeping the cluster-create and cluster-destroy steps always installed in the proper locations in the workflow.

Below is an example of a boundary-layer workflow that uses Dataproc resources:

name: my-dag-3

  start_date: '2018-10-01'
  project_id: my-gcp-project

- name: dataproc-cluster
  type: dataproc_cluster
    cluster_name: my-cluster
    region: us-east1
    num_workers: 128

- name: sensor
  type: gcs_object_sensor
    bucket: my-bucket
    object: my-object

- name: my-job-1
  type: dataproc_hadoop
  - dataproc-cluster
    main_class: com.etsy.foo.FooJob
- name: my-job-2
  type: dataproc_hadoop
  - dataproc-cluster
  - my-job-1
    main_class: com.etsy.bar.BarJob
- name: copy-data
  type: gcs_to_gcs
  - my-job-2
    source_bucket: my-bucket
    source_object: my-object
    dest_bucket: your-bucket

In this DAG, the gcs_object_sensor runs first, then the cluster is created, then the two hadoop jobs run in sequence, and then the job’s output is copied while the cluster is simultaneously deleted.

Of course, this is just a simple example; we have some complex workflows that manage multiple ephemeral clusters, with rich dependency relationships, all of which are automatically configured by boundary-layer. For example, see the figure below: this is a real workflow that runs some hadoop jobs on one cluster while running some ML training jobs in parallel on an external service, and then finally runs more hadoop jobs on a second cluster. The complexity of the dependencies between the training jobs and downstream jobs required boundary-layer to insert several flow-controloperators in order to ensure that the downstream jobs start only once all of the upstream dependencies are met.

Conversion from Oozie

One of our primary initial concerns was the need to be able to migrate our Oozie workflows to Airflow. This had to be an automated process, because we knew we would have to repeatedly convert workflows in order to keep them in-sync between our on-premise cluster and our GCP resources while we developed and built confidence in the new platform. The boundary-layer workflow format is not difficult to reconcile with Oozie’s native configuration formats, so boundary-layer is distributed with a parser that does this conversion automatically. We built tooling to incorporate the converter into our CI/CD processes, and for the duration of our cloud validation and migration period, we maintained perfect fidelity between on-premise Oozie and cloud-based Airflow DAGs.


A final requirement that we targeted in the development of boundary-layer is that it must be easy to add new types of operators, generators, or resources. It must not be difficult to modify or add to the operator schemas or the configuration settings for the resource and generator abstractions. After all, Airflow’s huge open-source community (including several Etsy engineers!) ensures that its list of supported operators is growing practically every day. In addition, we have our own proprietary set of operators for Etsy-specific purposes, and we must keep the configurations for these out of the public boundary-layer distribution. We satisfied these requirements via two design choices.

First, every operator, generator, or resource is represented by a single configuration file, and these files get packaged up with boundary-layer. Adding a new operator/generator/resource is accomplished simply by adding a new configuration file. Here is an example configuration, in this case for the AirflowBashOperator:

name: bash
operator_class: BashOperator
operator_class_module: airflow.operators.bash_operator
schema_extends: base

      type: string
      type: boolean
      type: object
        type: string
      type: string
  - bash_command
  additionalProperties: false

We use standard JSON Schemas to specify the parameters to the operator, and we use a basic single-inheritance model to centralize the specification of common parameters in theBaseOperator, as is done in the Airflow code itself.

Second, we implemented a plugin mechanism based on python’s setuptools entrypoints. All of our internal configurations are integrated into boundary-layer via plugins. We package a single default plugin with boundary layer that contains configurations for common open-source Airflow operators. Other plugins can be added by packaging them into separate python packages, as we have done internally with our Etsy-customized plugin. The plugin mechanism has grown to enable quite extensive workflow customizations, which we use at Etsy in order to enable the full suite of proprietary modifications used on our platform.


The boundary-layer project has been a big success for us. All of the nearly-100 workflows that we deploy to our production Airflow instances are written as boundary-layer configurations, and our deployment tools no longer even support python-based DAGs. Boundary-layer’s ability to validate workflow configurations and abstract away implementation details has enabled us to provide a self-service Airflow solution to our data scientists and engineers, without requiring much specialized knowledge of Airflow itself. Over 30 people have contributed to our internal Airflow workflow repository, with minimal process overhead (Jenkins is the only “person” who must approve pull requests), and without having deployed a single invalid DAG.

We are excited to release boundary-layer to the public, in hopes that other teams find it similarly useful. We are committed to supporting it and continuing to add new functionality, so drop us a github issue if you have any requests. And of course, we welcome community contributions as well!

No Comments

Double-bucketing in A/B Testing

Posted by on November 7, 2018 / 4 Comments

Previously, we’ve posted about the importance we put in Etsy’s experimentation systems for our decision-making process. In a continuation of that theme, this post will dive deep into an interesting edge case we discovered.

We ran an A/B test which required a 5% control variant and 95% treatment variant rather than the typical split of 50% for control and treatment variants.  Based on the nature of this particular A/B test, we expected a positive change for conversion rate, which is the percent of users that make a purchase.

At the conclusion of the A/B test, we had some unexpected results. Our A/B testing tool, Catapult, showed the treatment variant “losing” to the control variant.  Catapult was showing a negative change in conversion rate when we’d expect a positive rate of change.

Due to these unexpected negative results, the Data Analyst team investigated why this was happening. This quote summarizes their findings

The control variant “benefited” from double-bucketing because given its small size (5% of traffic), receiving an infusion of highly engaged browsers from the treatment provided an outsized lift on its aggregate performance.

With the double-bucketed browsers excluded, the true conversion rate of change is positive which is the results that we expected from the A/B test.  Just 0.02% of the total browsers in the A/B test were double-bucketed. This small percentage of the total browsers had a large impact on the A/B test results.  This post will cover the details of why that occurred.

Definition of Double-bucketing

So what exactly is double-bucketing?

In an A/B test, a user is shown either the control or treatment experience. The process to determine which variant the user falls into is called ‘bucketing’. Normally, a user experiences only the control or only the treatment; however in this A/B test, there was a tiny percentage of users who experienced both variants. We call this error in bucketing ‘double-bucketing’.

Typical user 50/50 bucketing for an A/B test puts ½ of the users into the control variant and ½ into the treatment variant. Those users stay in their bucketed variant. We calculate metrics and run statistical tests by summing all the data for the users in each variant.

However, the double-bucketing error we discovered would place the last 2 users in both control and treatment variants, as shown below. Now those users’ data is counted in both variants for statistics on all metrics in the experiment.

How browsers are bucketed

Before discussing the cases of double-bucketing that we found, it helps to have a high-level understanding of how A/B test bucketing works at Etsy.

For etsy.com web requests, we use an unique identifier from the user’s browser cookie which we refer to as “browser id”.  Using the string value from the cookie, our clickstream data logic, named EventPipe, sets the browser id property on each event.

Bucketing is determined by a hash. First we concatenate the name of the A/B test and the browser id.  The name of the A/B test is referred to as the “configuration flag”. That string is hashed using SHA-256 and then converted to an integer between 0 and 99. For a 50% A/B test, if the value is < 50, the browser is bucketed into the treatment variant. Otherwise, the browser is in the control variant.  Because the hashing function is deterministic, the user should be bucketed into the same variant of an experiment as long as the browser cookie remains the same.

EventPipe adds the configuration flag and bucketed variant information to the “ab” property on events.

For an A/B test’s statistics in Catapult, we filter by the configuration flag and then group by the variant.

This bucketing logic is consistent and has worked well for our A/B testing for years.  Although occasionally some experiments wound up with small numbers of double-bucketed users, we didn’t detect a significant impact until this particular A/B test with a 5% control.

Some Example Numbers (fuzzy math)

We’ll use some example numbers with some fuzzy math to understand how the conversion rate was effected so much by only 0.02% double-bucketed browsers.

For most A/B tests, we do 50/50 bucketing between the control variant and treatment variants. For this A/B test, we did a 5% control which puts 95% in the treatment.

If we start with 1M browsers, our 50% A/B test has 500K browsers in both control and treatment variants. Our 5% control A/B test has 50K browsers in the control variant and 950K in the treatment variant.

Let’s assume a 10% conversion rate for easy math. For the 50% A/B test, we have 50K converted browsers in both the control and treatment variant. Our 5% control A/B test has 5K converted browsers in the control variant and 95K in the treatment variant.

For the next step, let’s assume 1% of the converting browsers are double-bucketed. When we add the double-bucketed browsers from the opposite variant to both the numerator and denominator, we get a new conversion rate. For our 50% A/B test, that is 50,500 converted browsers in both the control and treatment variants. The new conversion rate is slightly off from the expected conversion rate but only by 0.1%.

For our 5% control A/B test, the treatment variant’s number of converted browsers only increased by 50 browsers from 95,000 to 95,050. The treatment variant’s new conversion rate still rounds to the expected 10%.

But for our 5% control A/B test, the control variant’s number of converted browsers jumps from 5000 to 5950 browsers. This causes a huge change in the control variant’s conversion rate – from 10% to 12% – while the treatment variant’s conversion rate was unchanged.

Cases of Double-bucketing

Once we understood that double-bucketing was causing these unexpected results, we started digging into what cases led to double-bucketing of individual browsers. We found two main cases. Since conversion rates were being affected, unsurprisingly both cases involved checkout.

Checkout from new device

When browsing etsy.com while signed out, you can add listings to your cart.

Once you click the “Proceed to checkout” button, you are prompted to sign in. You get a sign in screen similar to this.

After you sign in, if we have never seen your browser before, then we email you a security alert that you’ve been signed in from a new device. This is a wise security practice and pretty standard across the internet.

Many years ago, we were doing A/B testing on emails which were all sent from offline jobs. Gearman is our framework for running offline jobs based on http://gearman.org. In Gearman, we have no access to cookies and thus cannot get the browser id, but we do have the email address. So override logic was added deep in email template logic to bucket by email address rather than by browser id.

This worked perfectly. But the security email isn’t sent from Gearman; it is coming from the sign in request. So now our bucketing for the same browser id has this different bucketing based on email address rather than browser id.

This worked perfectly for A/B testing in emails sent by Gearman, but the logic applied to all emails, not just those sent by Gearman. Even though the security email is sent by the sign in request (not Gearman), the logic updated the bucketing ID to be the user’s email address rather than the browser id so that the browser might be bucketed into two different variants (once using the browser id and once using the email address).

Since we are no longer using that email system for A/B testing, we were able to simply remove the override call.

Pattern Checkout

Pattern is Etsy’s tool that sellers use to create personalized, separate websites for their businesses.  Pattern shops allow listings to be added to your cart while on the shop’s patternbyetsy.com domain.

The checkout occurs on etsy.com domain instead of the patternbyetsy.com domain. Since the value from the user’s browser cookie is what we bucket on and we cannot share cookies across domains, we have two different hashes used for bucketing.

In order to attribute conversions to Pattern, we have logic to override the browser id with the value from the patternbyetsy.com cookie during the checkout process on etsy.com. This override logic works for attributing conversions; however during sign in some bucketing happens prior to the execution of the override logic by the controllers.

For this case, we chose to remove bucketing data for Pattern visits as this override caused the bucketing logic to put the same user into both the control and treatment variants.


Here is a dashboard of double-bucketed browsers per day that helped us track our fixes of double-bucketing.