About the RoleJoin an early-stage AI/ML developer tooling team as its first full-time engineer outside the founding group. You will build systems that detect failures in AI agent conversations and generate verified fixes, helping customers ship more reliable software.
What You'll Do- Build end-to-end product capabilities, from data ingestion and failure detection to patch generation and verification.
- Design reproduction logic, guardrails, and checks that help prevent unsafe or incorrect changes from reaching customer repositories.
- Turn customer problems into shipped solutions, working directly with customers to understand traces, production failures, and requirements.
- Make architectural decisions across schemas, service boundaries, and system tradeoffs.
- Establish engineering practices for testing, observability, on-call, and releases.
- Work across data pipelines, detection systems, and user-facing review experiences as product needs evolve.
What We're Looking For- At least 3 years of experience shipping production software, with a track record of contributing beyond your core specialty.
- Strong full-stack TypeScript skills, including experience building user-facing products and core services.
- Experience building and operating AI agents in production, with an understanding of agent failure modes and debugging.
- Experience designing and optimizing high-volume data pipelines with attention to cost, observability, monitoring, and production debugging.
- Ability to independently take loosely defined customer problems through architectural decisions and delivery.
- Comfort working directly with customers and in an early-stage team with broad ownership.
Compensation & BenefitsThe annual salary range is $150,000 to $200,000. Visa sponsorship is not available.
LocationOn-site in San Francisco, California. The role requires working from the office six days per week.
