
Posted today
Applied Machine Learning Engineer
AI Summary
Develops tools, harnesses, and evaluations that let frontier AI models control and reason about real-world robots, testing improvements on a robot fleet and analyzing model failures.
About this role
Pantograph is building general purpose intelligent robots that can safely and autonomously accomplish useful work in the real world. We believe the path to capable robots runs through general intelligence rather than narrow, robot-specific skills, and that there is an enormous amount of work worth doing in the physical world that would improve quality of life for people. As AI models make it easier to build software around our needs, we hope robots can empower people to shape the physical world.
Across research, hardware, and software, we focus on finding simple ideas that work, executing them extremely well, and scaling them up. We value clear thinking, careful execution, and systems we can understand, trust, and improve.
We're looking for an applied machine learning engineer to help turn increasingly capable AI models into increasingly capable robots. You'll develop tools, interfaces, and features that frontier models can use effectively to accomplish real-world tasks. That means understanding what models need to observe, reason, and act, identifying where performance breaks down, and testing improvements on our robot fleet.
The work includes building harnesses that connect models to robot observations and actions, designing useful tools, improving context and memory, and developing evaluations that tell us whether we're making progress. You'll explore different ways of organizing robot intelligence: when a frontier model should control a robot directly, when it should coordinate faster, specialized policies, and when it should write its own tools or control code. You'll evaluate the tradeoffs in capability, reliability, latency, and how easily we can understand and supervise the system.
You'll work closely with our research, robot software, and product teams, building a tight loop between experiments and experience on real hardware. We're looking for someone who combines excellent experimental judgment with a strong intuition for model behavior, and enjoys building things to test their ideas. You'll investigate whether a failure calls for a better observation, a different tool, a change in orchestration, or a fix elsewhere in the system.
You might be a good fit if you:
Are excellent at designing experiments and interpreting results: separating competing explanations, identifying confounders, and knowing what the evidence does and doesn’t establish
Have a strong intuition for how models behave: what they understand, what confuses them, and how their behavior changes with the tools, context, and feedback they receive
Have built and improved AI systems that people use, with clear evidence of what made them better
Are comfortable moving between model APIs, data pipelines, inference, and application software
Enjoy digging into failures and understanding how models behave beyond aggregate metrics
Can build reliable software around models, including tools, harnesses, and testing infrastructure
Are pragmatic about methods and care about what works on real tasks
Communicate clearly and enjoy working across research and engineering
Use AI tools extremely effectively
Bonus if you've worked with:
Agent harnesses, tool use, memory, or long-horizon tasks
Systems where models write and execute code or develop their own tools
Classical robotics, including controls, motion planning, or perception
Robot learning, imitation learning, or reinforcement learning
Evaluation design, failure analysis, or safety testing
Inference optimization, including latency, throughput, and cost
Learning from fleet data or operating ML systems across varied real-world environments
We care much more about what you've built and learned than credentials. We're a small, fast-moving team working together in person in San Francisco. If this sounds exciting to you, we want to hear from you.
Skills
Explore related jobs
More jobs at Pantograph
Similar Agent Harnesses jobs
- S
Software Engineer, Agent HarnessSim · San Francisco - L
Software Engineer, Agent HarnessLm Studio · New York City - E
AI Software Engineer, Agent HarnessEnCharge AI · India - C
Product Manager, Agent Harness & ModellingCohere · Toronto - V
Software Engineer - Agent HarnessValerie Health · San Francisco, California - S
Senior Agent Harness EngineerSavvy · NYC Office
Jobs in San Francisco
- O
Integrated Marketing ManagerOpenAI · San Francisco - V
Director, Partner MarketingVercel · Hybrid - San Francisco - P
FP&A Manager, Commercial & MarketingPlanet · San Francisco, California - E
Office ManagerEnvironmental Science Associates · San Francisco, California - C
Specialist, Strategic Partnerships (Payments)Checkout.com · San Francisco - L
Account Manager, Lyft AdsLyft · San Francisco, California
Browse these categories
Market data for ai / ml engineer roles
All reports →- SeriesRole reportsOne role family at a time: how many openings, what changed this week, who is hiring, what it pays.
- SeriesSalary reportsWhat employers publish in job postings, by level and workplace. Not self-reported pay.
- Market overviewState of tech hiring, September 2026: up 4.8%Tech hiring rose 4.8% month over month in September 2026, with 411,122 new listings. Customer support and account executive roles led the growth.