Agi Inc
Research Engineer - Evals
San Francisco Office. Posted 5 months ago
The role
You will build the eval harness for AGI across model capability, agentic behavior, on-device performance, and end-user experience. You will own eval suites, dashboards, and tooling to gate model and agent releases.
What you will do
- Own eval suites that gate every model and agent release
- Develop dashboards and tooling for researcher experiment loops and leadership decisions
- Set the bar for what counts as ready to ship
- Assist research by ensuring measurements align with goals
- Assist product engineers by instrumenting real-user behavior on real devices
- Assist partnerships by translating performance into language for OEM partners
Required qualifications
- Send a link to an eval, benchmark, or measurement system you built
- Provide one paragraph on what decision that system changed
- Provide resume or LinkedIn
Benefits
- Competitive cash
- Meaningful equity
- Top-tier relocation support
- Top-tier immigration support
About the company
agi-inc is a stealth team of elite founders and AI researchers building everyday AGI. They are industry leaders in mobile and computer-use agents.