An early-stage AI infrastructure company is building the continuous improvement infrastructure for enterprise agents.
They create a closed loop between how agents perform in production and the knowledge they rely on, identifying what went wrong, correcting the underlying cause, and continuously improving both the knowledge and the agents using it. Every failure becomes a signal that makes the entire system more reliable over time.
We are looking for a backend-leaning Full-Stack Engineer to help build the infrastructure behind the current platform and the next generation of agent products. Most of the difficult work is in the backend: data modeling, queues, execution pipelines, integrations, observability, reliability, and debugging distributed workflows. You should still be comfortable taking a feature through the full stack and shipping the product surface that makes the underlying system useful. You will have meaningful ownership over architecture and product decisions while working directly with a small, highly technical team and enterprise customers.
What You'll Work On
Build the backend systems powering the agent evaluation and continuous-improvement platform.
Improve the data models behind evaluation datasets, experiment runs, agent traces, knowledge versions, findings, and customer-defined quality criteria.
Own enterprise integrations throughout their entire lifecycle: authentication, OAuth, webhooks, incremental syncs, backfills, rate limits, schema changes, retries, monitoring, and production support.
Own features from technical design and implementation through deployment, monitoring, customer feedback, and iteration.
Work directly with customers when needed to understand their systems, diagnose failures, and convert one-off requirements into reusable product capabilities.
- 2-4 years of experience as a backend-leaning full-stack software engineer
- Strong skills in Python and TypeScript; LLM experience is a plus but not required
- Experience building production systems for enterprise customers
- Experience at an early-stage startup (e.g., under 50 people) with real ownership
- Full-stack experience, comfortable with both frontend and backend (Python preferred)
- Professional experience building with TypeScript
- Scrappy, entrepreneurial mindset; founder or hackathon background is a plus
- Comfortable with the fast-paced, evolving nature of a startup
- Bachelor's degree in Computer Science or a related field
- Prior agent eval, LLM, or evals/QA experience is a plus