Applied Researcher - Computer Use

Highlights

• Build long-horizon computer-use environments from real partner workflows.
• Train agents to recover from interface changes and partial failures.
• Develop evaluations that distinguish genuine completion from plausible behavior.

What they offer

• Equity incentives
• Flexible work
• Visa sponsorship & relocation support
• Professional development budget
• Team off-sites & conference attendance

Ideal candidate

Experimental builder who thrives in ambiguity and partner-facing research, turning messy real-world workflows into rigorous evaluations and training loops that make agents reliably useful.

Requirements

• Relevant experience in multimodal agents, computer use, browser or desktop automation, RL, imitation learning, HCI, embodied agents, or long-horizon agent evaluation
• Hands-on experience building and debugging model-training, post-training, agent, or evaluation experimental loops
• Strong experimental judgment with meaningful baselines, careful trace inspection, and clear failure attribution
• Strong Python and ability to build artifacts others can run, inspect, and extend
• Comfort owning ambiguous problems and keeping partners and teammates current