The role
- COMP
- $130K - $200K
- EQUITY
- ~1%
- LOCATION
- Washington DC · Canada
- WORKPLACE
- Remote
- EXPERIENCE
- 1+ years
- VISA
- None, Visa transfers
- STACK
- Python
- INDUSTRY
- AI, Devtools
/roles — ROLE_479
AI Research Scientist
Early-stage AI infrastructure startup helping enterprises build agents that behave reliably in production
The company
An early-stage startup building a platform that helps enterprises make AI agents dependable in production through evaluation, simulation, and ongoing optimization of agent behavior.
- STAGE
- Series A-stage
- FUNDING
- $6M raised
- TEAM
- founding team of <10
- FOUNDED
- 2024
- BACKING
- VC-backed
JD — the work
About the role
This role is research-heavy and suits a person who moves easily between ideas, experiments, and product. You would develop the core methods behind a platform that helps enterprises take AI agents from fragile demos to workflows that hold up in the real world, and you would help steer both the research agenda and the product. The team is small, about ten people, and works remotely across the US and Canada.
What you'll do
- Improve the algorithms and systems that make agents reliable at every stage, from design to deployment
- Invent new ways to simulate, evaluate, and optimize how agents behave in realistic settings
- Carry promising ideas through to prototypes and then to features customers rely on
- Work with engineers and product leads to turn research progress into value customers can feel
- Shape the company's technical roadmap for how agents are tested, tuned, and made reliable
What they're looking for
- 1+ years of hands-on research or engineering in modern AI
- Strong Python and experience running AI experiments from idea to result
- Comfort switching between open-ended research and practical product work
- Original ideas about agentic AI and the drive to see them shipped