Careers

We're building the user model: one model that predicts your tests and finds customer friction. If pushing the frontier of user simulation sounds like your kind of problem, we'd love to hear from you.

ResearchSan Francisco / RemoteFull-time

Founding ML Researcher, Finetuning

Own the finetuning stack behind Jurny's user model. You'll design the training pipelines, data curation, and post-training methods that turn frontier base models into faithful simulators of real user behavior.

What you'll do

Lead the design and execution of our supervised and reinforcement finetuning pipelines end to end.
Build data curation, filtering, and synthetic-data strategies that improve model fidelity on real user behavior.
Run rigorous evaluations and ablations to understand what actually moves prediction quality.
Set the technical direction for post-training as one of the first research hires.

Apply for this role

ResearchSan Francisco / RemoteFull-time

ML Researcher, Human Likeness

Make simulated users feel human. You'll research the behavioral realism of our user model, how it hesitates, gets confused, drops off, and decides, so predictions match how real people actually behave.

What you'll do

Define and build evaluations for behavioral realism and human-likeness of simulated users.
Research modeling approaches for persona grounding, intent, and realistic friction and drop-off.
Partner with the finetuning team to translate behavioral objectives into training signals.
Turn messy human-behavior data into benchmarks the whole team can optimize against.

Apply for this role

Don't see the right role? We're always looking for exceptional people. Reach out at careers@usejurny.com.