: ML Feature Management Tech Lead
141 Portland St, 6th Floor
Cambridge, MA 02139

Who we are:

Cortex empowers internal teams to efficiently leverage ML by providing a platform and by unifying, educating, and advancing the state of the art in ML technologies within Twitter.

We win when our customers win by helping our users stay informed, share and discuss what matters; by serving the public conversation. We’re building an AI-first company and every major initiative is increasingly dependent on the successful application of machine learning. Cortex is at the nexus of this evolution. Our team of ML software engineers is constructing one of the strongest machine learning platforms in the world, based on the latest ML industry practices, deep learning, engineering excellence, and powered by Twitter data at scale.


The ML Feature Management (MLFM) team is part of the Cortex Platform group and, together with the other four Cortex Platform teams, develops tools and infrastructure that standardizes machine learning development at Twitter on a common platform so ML engineers in our customer teams can do their work better and faster and with a more modern stack that follows broad industry trends.


More specifically, our team’s vision is to maximize the velocity of Feature Engineering. To date we’ve built and heavily invested in Twitter’s Feature Store, to share features across different ML product teams in production and thereby to leverage feature engineering investments across the company. This has been a wildly successful mission over the last two years and all ML teams have been migrating to the Feature Store which is expected to complete in 2020.


The next frontier is to improve the tooling across the entire feature lifecycle, from the experimental stage in interactive notebooks with GCP tools, to easy productionization and A/B testing of winning candidates, and arcing all the way to deprecating and removing expired features. Some examples of tools we expect to be prototyped along the way are: statistical methods for opportunity sizing feature ideas; Feature Store notebook integration; automatic estimation of capacity requirements and dollar cost of deploying a new feature to production; easy hooks to run A/B tests for new features; tools for analysing feature importance and identifying features for deprecation that aren’t pulling their weight. 


In the shorter term we expect this will entail prototyping a string of product ideas, to be validated with pilot partners for a few quarters. Winning prototypes will then be developed into mature products. In the longer term, and with proper meta-data instrumentation in place on tools, feature and model usage, this could lead to model-based semi-automated feature recommendations.


What you will do:

We’re looking for a team member to join us in tech lead capacity to head up this effort. Across the industry to date, feature engineering workflows remain fragmented and employ highly bespoke tools. Unifying and consolidating the entire feature lifecycle tool chain and thereby establishing best practices for feature engineers and modelers to achieve order-of-magnitude gains in productivity is an industry-leading opportunity. The right candidate can turn this into a company-wide competitive advantage.


Since this project is brand new and there are no industry peers to follow that have published end-to-end solutions, we expect a fair amount of uncertainty and ambiguity. The ideal person will not only be comfortable with ambiguity but view it as an exciting opportunity to quickly explore a multitude of options. The ideal candidate will have the knowledge and experience to build out validated ideas into full-fledged products for our ML customer teams. The ideal candidate will keep a portfolio of product ideas at different stages of maturity in flight and produce a steady cadence of robust product innovation.


Who you are:

To be successful, we believe the ideal candidate will have the following traits:

  • Strong customer focus.
  • A passion for machine learning.
  • Ability to bring partners together across organizational and functional boundaries.
  • Ability to articulate a clear vision and enroll the team and partners into it.
  • The technical strength and rigor in process to lead by example.
  • A team-focused mindset to inspire and grow contributors on and off the team.
  • A growth mindset that views failures (e.g. of hypotheses or prototypes) as an effective way to learn and iterate.
  • A multiplicative effect on team members across different levels of seniority, skills and geographical boundaries.
  • The organizational skills to keep an emerging product effort with lots of uncertainties on track.
  • Motivated by delivering impactful products that accelerate the feature engineering efforts of our customers.


By nature of the problem domain, we expect successful candidates to have experience in:

  • Building and delivering working software through an iterative, agile process.
  • 2+ years experience with ML problems and tools either through first-hand modeling or close collaboration with modeling engineers or data scientists.
  • 2+ years in Sr engineering capacity with demonstrated leadership skills.
  • 5+ years work experience in software engineering in the areas of distributed data processing, developer tooling and/or ML platform space.
  • M.S. or Ph.D. degree in computer science or a related field or equivalent work experience.


The position is available in New York but we're also open to Seattle or other locations.




Hiring Process

Step 1

After you apply, a recruiter may reach out to you for an introductory call.

Step 2

If your background is a match for the role, you may phone interview with 1-2 people.

Step 3

If you continue through the process, you will come onsite 1-2 times to interview with a total of 5-10 people.