The MITRE Corporation
True Talent Solutions Partners
Full-Stack Engineer, Agent Evaluation & Reliability
TypeScriptNode.jsPostgreSQL
About the Role
Join a small, technical team building AI agent evaluation and reliability infrastructure for enterprise customers. This backend-leaning full-stack role focuses on making agent performance measurable and improving the systems that support reliable AI products.
What You'll Do
- Build backend systems and data models for evaluation datasets, experiment runs, agent traces, knowledge versions, and quality criteria.
- Develop and operate enterprise integrations, including authentication, data syncing, retries, monitoring, and production support.
- Ship features across the stack, from technical design and implementation through deployment, monitoring, and iteration.
- Work with customers to diagnose system failures and turn their needs into reusable product capabilities.
- Build data pipelines and interfaces for collecting, analyzing, and visualizing evaluation results.
What We're Looking For
- At least 3 years of experience building and shipping production systems as a full-stack engineer, ideally including work at a fast-paced startup.
- Hands-on machine learning model development experience, along with knowledge of evaluating and testing AI agents or other nondeterministic systems.
- Strong backend skills with TypeScript and Node.js or Bun, and experience building production frontend features.
- Strong PostgreSQL skills, including query optimization, and practical experience with Redis, job queues, background workers, or workflow systems.
- Experience designing APIs and backend services, building evaluation data pipelines, and developing dashboards or visualization interfaces.
- Familiarity with observability, monitoring, logging, and production troubleshooting.
Compensation & Benefits
- Base salary: $120,000 to $180,000 annually.
- Visa sponsorship is available.
Location
On-site in San Francisco, California, United States.
