17 hours ago
San Francisco, CA, USAMid Level / Senior
Base Salary
$150k - $300k/yr
Responsibilities
- Own the agent layer end-to-end, including context engineering, self-learning systems, evals, the agent harness, and the meta-harness.
- Build a compounding knowledge base of undocumented systems so future integrations benefit from prior learning.
- Design and ship self-healing workflows that autonomously detect, diagnose, and recover from production failures.
- Build and maintain a meta-harness for creating, evaluating, and evolving agents.
- Set patterns and standards for how agents are built, measured, and improved.
- Help grow the team and eventually lead the people responsible for agentic systems.
Requirements
- 2 to 9 years of professional software engineering experience with hands-on agentic software built within the last year.
- Strong computer science fundamentals and ideally a technical degree from a well-regarded university.
- Proficiency with TypeScript and experience working with LLMs, evaluations, and agent tracing in production.
- Experience building context or self-learning systems that accumulate knowledge to improve agent performance.
- A demonstrated record of shipping autonomous failure detection, diagnosis, or self-healing workflows to production.
- Experience with eval-driven iteration and structured context engineering.
- Startup, founding, YC, or top-tier venture-backed company experience is strongly preferred.
- Reusable agent harness or meta-harness experience is a plus.
Benefits
- Salary range of $150,000 to $300,000 USD annually.
- O-1 visa sponsorship is available for qualified candidates with a concrete relocation plan.
- On-site in San Francisco, California, five days per week.
- Candidates currently elsewhere in the United States may relocate with a concrete, dated relocation plan.