Senior AI Engineer

Join an early-stage AI company building software for physical commodity trading and operations. You will build reliable agent systems that turn modern models into useful workflows, from early experiments through production use.

• Build agent harnesses that provide models with the tools, context, and execution environments needed to complete tasks across the product and connected systems.
• Design agent planning, action-taking, state management, failure recovery, and handoffs to people.
• Prototype model-powered features against real customer problems and develop promising experiments into production capabilities.
• Create evaluations for task completion and decision quality, and turn production failures into reproducible test cases.
• Inspect execution traces and debug prompts, context, tool interfaces, orchestration, and model behavior.
• Improve agent autonomy using appropriate permissions, checkpoints, and human intervention.
• Evaluate models and approaches for capability, reliability, latency, and cost in real workflows.
• Build reusable tools and harness components that make features easier to develop and operate.

• Four to ten years of software engineering experience, including ownership of production systems through launch, operation, and debugging.
• Hands-on experience building LLM or agent systems used by real users, including orchestration or harnesses, tool calling, and evaluations.
• Experience measuring and improving production LLM quality, latency, or cost using evaluations, trace inspection, and performance analysis.
• Strong software engineering skills in addition to any data science background, and the ability to move from prototypes to reliable production features.
• Experience at an early-stage startup and good judgment about when agents should act autonomously or hand work to a person.
• Strong computer science fundamentals and experience building practical AI workflows or connected-system integrations.

Base salary: $200,000 to $250,000 USD annually. Visa sponsorship is not available.

Remote within the United States. Primary location: San Francisco, California.