Staff Research Engineer, Multi-Agent Scaling
Core
Design, run, and interpret large-scale experiments on multi-agent teams to study scaling laws, performance, and coordination bottlenecks while building the underlying platform and evaluation tooling.
Role type
Staff Research Engineer (Multi-Agent Scaling)
Builds
Platform and evaluation systems for running and measuring large agent teams
Domain
AI Safety, Multi-Agent Systems, Large Language Models (LLMs)
Deliverable
production ML models
Required skills
Software engineering, Machine Learning, Research engineering, Experimental design, System debugging, Quantitative analysis, Tooling development, Cross-functional collaboration
Preferred skills
Distributed systems, LLM evaluation/benchmarks, Agentic systems, Scaling laws, Operations research, Statistics, Economics, Physics
Technologies
LLMs, Distributed systems, Schedulers, Sandbox environments, RL infrastructure
Responsibilities
Design and run large-scale experiments on agent teams; Investigate performance/efficiency changes with team size and compute; Build and scale systems for large agent teams; Design evaluations for long-horizon problems; Build tooling and metrics for researchers; Partner with research teams to enable experiments
Seniority
Staff, hands-on IC with strategic impact
