About the company: Well-funded NYC AI startup building a behavior model that predicts how online shoppers decide. Its AI runs live experiments on eCommerce stores to lift conversion. Small team, direct work with the founders, 4 days/week in Midtown.
Why join
Every day, millions of people decide whether to buy something online, and the reasons behind those decisions are still mostly guesswork. The company is working to quantify human psychology: to build an embedding model for human behavior that understands how people react to a page, a price, or a single sentence, and can predict what will work before anyone runs a test.
Every experiment the AI runs on a live store teaches that model something about how people decide. The goal is for any company to be able to ask, "I'm selling product X to audience Y, how do I optimize for Z?" and get an answer it can trust.
The company is early and well funded. The people who join now will pick the tools, design the models, and help choose who joins next.
Senior Fullstack Engineer
Tens of millions of shopper events flow through the systems every day, and every experiment the AI runs adds another piece of evidence about why people buy. You'll build the systems that capture that evidence, turn it into something a model can learn from, and put the model's ideas in front of real shoppers without breaking anything. It's hands-on work across the whole stack, on a small team where you'll design most of what you build.
What you'll work on
- The data behind the model. The pipeline (AWS, ClickHouse) takes in tens of millions of web events a day. You'll make it faster and more reliable, and shape it into data a behavior model can learn from.
- The model itself. You'll help decide how to represent shoppers, pages, and changes so the model can tell what will work before it's tested.
- From idea to live test. When the AI proposes a change, your code turns it into a working version of the page, checks that it doesn't break checkout or tracking, and serves it to real shoppers.
- What clients see. The dashboard where clients review what the AI wants to try, approve it, and follow the results.
Stack: mostly Python on serverless AWS, JavaScript on the frontend, and Postgres alongside ClickHouse.
You'll love this job if
- You want to work on an open research problem and still ship to production every week.
- You like moving across the stack, from a ClickHouse query to a dashboard screen, in the same week.
- You'd rather put a rough version in front of real data in three days than polish a design for a month.
- You get excited when a result surprises you and you have to figure out why.
What you bring
- 5+ years of experience
- You've built and run production systems end to end, and you've fixed them when they broke. Python and JavaScript are the main languages, and you can pick up whichever one you know less.
- You've worked with data at scale: event pipelines, streaming, or analytics databases.
- You've led a major project or a small team, from the design decisions to the launch.
- You've built with LLMs or other ML models, and you know where they fail.
- Bonus: you've trained or deployed ML models in production, or you've built a product from zero at a startup.
Values
- Curiosity and excellence: Asking questions is the job. The team pushes on each other's ideas, and a failed experiment teaches as much as a winning one.
- Whatever it takes: If something needs doing, whoever can do it does it, whatever their title.
- Ownership: Shipping something is where the work starts. You follow it all the way to the result.
- Agility: Most ideas go from whiteboard to working prototype in two or three days.
- Enjoy the ride: Building something new comes with rough weeks. The team gets through them together.
What's offered
- Base salary $150K - $210K plus meaningful equity
- Medical, dental, and vision for you and your family
- Unlimited PTO
- Tax-free commuter and parking benefits
- Office-first culture (4 days/week) in Midtown Manhattan
- Work directly with the founding team
- Free One Medical membership
- 401k