Skip to main content

Axiombio

Agent Harness Engineer

  • Confirmed live in the last 24 hours
  • No salary listed
  • Mid level
  • Full-time
  • On-site · SF Global HQ
  • Added 2 months ago

About this role

Axiom is seeking a skilled Agent Harness Engineer to build and maintain the infrastructure that transforms cutting-edge AI models into robust, reliable agents for scientific analysis. You'll own the entire lifecycle, from data pipelines and execution environments to evaluation systems and observability tooling, ensuring the reliability and effectiveness of these agents in the drug development process. This role requires a proactive individual who thrives in a rapidly evolving environment and is passionate about translating expert knowledge into measurable systems.

What you'll do

  • Develop and maintain the agent harness, including scaffolding and tooling.
  • Build data pipelines and storage systems for runtime context and training data.
  • Design and implement secure, reproducible execution environments.
  • Create evaluation systems, including offline suites and LLM-as-judge pipelines.
  • Develop observability tooling to monitor and debug agent performance.
  • Engineer context management for long-horizon agent runs, ensuring coherence and recovery.

What they're looking for

  • Python
  • Modal
  • DuckDB
  • FastAPI
  • Docker
  • Containerization
  • Terraform
  • LLM APIs
Apply with Autofill

Opens the application — the Jobs AI extension fills it for you. Set up autofill

Opens the official application on the employer’s site. No login required.

Axiombio

View all jobs at Axiombio

Likely interview questions

  • Describe your experience building agentic systems, including challenges faced and solutions implemented.
  • How do you approach designing and implementing evaluation systems for complex AI agents?