Data Platform Engineer

Description

Analytical Flavor Systems is a venture-backed startup that models human sensory perception of flavor, aroma, and texture using proprietary machine learning in order to predict consumer preference of food and beverage products. The work we do allows our clients in the food & beverage industry to ask and answer questions about:

• their competitive landscape ("what do people like and dislike about my competitors' products?")
• optimizing existing products ("how can I make this cookie taste better?")
• novel flavor combinations for new product development ("would people like it if I combined matcha and strawberry in yogurt?")

Our data science capabilities are evolving from report generation to a data platform. That's where you come in.

The work expected of a data platform engineer at Analytical Flavor Systems covers several major areas:

• Building out a data model on Datomic to capture the information generated by our day-to-day data processing tasks in an immutable and readily queryable data store.
• Creating a data platform application layer to serve the needs of our data science team, our web console and our mobile data collection app.
• Rewriting existing data science code for execution in a distributed rather than single-machine environment.
• Maintaining and enhancing batch processing jobs to make them faster, more reliable, and more observable.
• Refactoring an existing codebase to be more modular: separating data transformations, modeling, and prediction steps into discrete functions with well-understood inputs and outputs, while testing for regressions in predictive capabilities.

Depending on your background and areas of expertise, your day-to-day work may focus more on one of these areas than others, but you should be able to keep the big picture in mind, and understand how the changes you make to one part of our system affect the whole. Your work will improve our ability to execute this code reliably, and replicate previous results. This work will also help us observe and capture the outputs of the analytical operations we perform so we get better insight into the state of the systems built atop our data science code.

You will be expected to become comfortable working in both Clojure and R, though no prior experience in R is required. This role offers you the chance to help develop the language of our research domain, which may help us identify potential new avenues of theoretical research in human sensory perception.

We are only considering candidates with USA work authorization or work visa (including OPT). AFS can sponsor H1-B renewals or transfers.

Requirements

Candidates should have at least 4 years of total programming experience, with at least 1 year of work in either a data engineering context or in building backend systems. Experience with Clojure or other functional programming languages is a plus, but not a requirement. Functional programming is as much a style and idiom of development as it is a family of languages. Candidates that have experience building modular systems that put data front and center, regardless of the implementation language, should be attracted to this role.

Candidates with experience supporting the work of researchers and data scientists are also strongly encouraged to apply. Have you made an analytical method production-ready after reading through someone else's prototype code? Are you interested in interoperability between R and Clojure? Have you helped deploy and monitor models in production? Experience with these questions gives you a good understanding of the requirements and scope of the systems we build.

The company is roughly 15 people total, so candidates will be working closely with other teams and areas of the business. Good communication skills, especially across varying levels of technical depth and skill, are preferred.

A good candidate should have experience in at least two of the following areas:

• Data science and analytics: you have enabled more powerful access to data for both technical and non-technical stakeholders. You understand how to support and enhance systems based on machine learning, and aren't afraid of diving in to build a more efficient implementation of an algorithm than one provided by a library.
• Data modeling: you have, either on your own or as part of a team, designed or extended a relational database schema to support application and business requirements.
• System design and maintenance: you know how to build and extend existing systems to make them more observable, fault-tolerant, and performant. You can ssh into a remote box to contextualize a problem that doesn't have an obvious cause.
• Automated QA and testing: you know what the invariant properties of both individual functions and system components are, and can represent those properties in code.

Benefits

• Competitive salary with offers starting at $120K
• Standard benefits package (health insurance/vision/dental + 401k)
• Equity stake (Restricted Stock Units with 4-year annual vesting schedule)
• Remote-friendly (who isn't these days?). While we do plan on an eventual return to our office space in Manhattan once it's safe, immediate relocation to the NYC area is not an expectation of this role.
• That said, if you do end up in our NYC office you'll be able to join regular in-person tasting panels to get hands-on experience with the sensory data collection methods we use.
• Unlimited vacation policy
• Professional development budget

Type: Full-time, Remote.