Software Engineer — LLM Agents & Automation Systems
SF Bay Area | On-site
We’re representing a frontier AI research group building autonomous systems that can reason, plan, and execute real engineering workflows.
If you’ve worked on LLM applications, tool-calling systems, workflow engines, automation platforms, or reliability-heavy backend systems — this role will make sense, even if you’ve never held the title “Agent Systems Engineer.”
⭐ The Opportunity
Join a small, high-performing team building the systems that let AI agents reliably:
✅ break down complex tasks
✅ choose the right tools
✅ execute multi-step workflows
✅ recover from failures
✅ deliver consistent outcomes (not just good demos)
This is the engineering layer that turns LLMs into real autonomous software.
What You’ll Do
• Build the workflow + orchestration layer that powers LLM-driven agents
• Design structured interfaces for tool use, actions, and task execution
• Implement reliability features like retries, rollback, checkpointing, and reproducibility
• Connect agents to real-world engineering tools, APIs, and data systems
• Work closely with ML + infra teams to ship production-grade autonomy
What We’re Looking For
You might be a great fit if you’ve built things like:
• LLM tool-calling / function-calling systems
• Workflow engines, orchestrators, automation platforms
• Backend systems where correctness + failure handling really matter
• Structured pipelines (state machines, planners, evaluators, task runners)
• Strong engineering fundamentals + comfort in fast-paced environments are key.
Why This Role Excites People
• Build autonomy that impacts real workflows (not just prototypes)
• High ownership over architecture + system design
• Work alongside researchers + engineers pushing the frontier
• Competitive salary + strong equity upside
