About the Role
This is a founding-level engineering role owning the agent layer end-to-end at a gig economy and flexible staffing platform. You'll work directly with the CTO to build production agentic systems that do real work with real consequences — setting the patterns every future agent at the company will follow. It's an early and high-ownership position at the frontier of applied AI engineering.
What You'll Do
-
Own the full agent layer: context engineering, self-learning systems, evals, the agent harness, and the meta-harness.
-
Build a compounding knowledge base of undocumented systems so each new integration is better than the last.
-
Develop self-healing workflows — agents that detect production failures, diagnose them, and recover autonomously.
-
Build and maintain the harness agents run on, plus the layer that enables building, evaluating, and evolving new agents in days.
-
Set the engineering patterns for how every agent across the platform gets built, measured, and improved.
-
Help hire and eventually lead the team responsible for agentic systems.
-
Work as a peer to the founding team and experiment relentlessly.
What We're Looking For
-
2–9 years of software engineering experience, with hands-on agentic software built within the last year.
-
Deep familiarity with harness-centric agent building: context engineering and eval-driven iteration.
-
Strong understanding of agent tracing and evals.
-
Implementation-focused engineer who ships working systems in production — not a researcher.
-
Proficiency in TypeScript; experience with Effect-TS, PostgreSQL, and LLM/agent tooling is a strong plus.
-
Background from high-calibre engineering environments is valued.
Compensation & Benefits
-
Salary: $150,000 – $300,000 USD annually
-
Visa sponsorship: Available
Location
On-site in San Francisco, CA, United States.