Where multiple locations are listed for this role, the position may be based in any of those locations, with priority determined according to the order of listing.

Agents generate an unusual amount of data: every run, every step, every recovery from failure. Buried in that data is the answer to which agent behaviors actually work and where users get stuck. Not enough of it is being read today.

You'll own analysis that shapes what we build next, working directly with product and engineering.

What you'll do

Build and maintain dashboards that track product health: activation, retention, task success, and agent reliability

Analyze agent run data to find where tasks fail, stall, or need human rescue

Answer product questions with data, and be honest when the data can't answer them

Help define and instrument the events and metrics we should be tracking in the first place

Support experiment design and read out results without overclaiming

Turn analysis into recommendations, with a clear view of what you'd do next

Make your work reproducible so others can build on it

You might be a fit if

You're pursuing a Bachelor's or Master's degree in Statistics, Computer Science, Economics, Engineering, or a related field, or equivalent experience

You're solid with SQL and comfortable in Python or R for analysis

You've built a dashboard or analysis someone else actually used

You understand the basics of experimentation and are appropriately skeptical of noisy results

You can explain a finding to someone who won't read the notebook

You're curious about the product, not just the numbers

Bonus: experience with product analytics tools, dbt, or working with event data at scale