Senior Full-Stack Engineer

Most of what makes this product work is AI: pipelines that read messy supplier documents and turn them into structured compliance data, the agents and retrieval behind them, and the evals that keep them trustworthy. The work is varied. One week you might be building a news-scraping agent that tracks regulatory changes at scale; the next, a voice agent that calls suppliers to chase down missing documentation. The core product is already in customers’ hands, and our focus now is taking it to the next level. The hardest of that work is on the AI side, where a lot of the problems don’t have settled answers yet.

We’re a small, senior team that ships with AI coding tools every day, and we expect you to work the same way.

What you’ll do

• Help design features, not just build them: shape scope, UX, and technical approach alongside the team from the start.
• Ship new AI capabilities end-to-end: design an agent or extraction pipeline, get it working, prove it out with evals, and put it in front of customers.
• Push the systems further: better accuracy, lower cost, more autonomy, new document and data types we don’t handle yet.
• Take on open-ended problems like the news-scraping and voice agents above, where the right approach isn’t obvious going in.
• Build the backend and product surface these systems need: APIs, data models, services, and the occasional UI.
• Raise the bar on quality as we grow, from evals to tests to how we review AI-written code.

• 5+ years building production web apps in Node.js and TypeScript.
• Proven frontend experience with modern web frameworks (React preferred)
• Microservices experience: design, operation, and failure modes (retries, idempotency, observability).
• Experience with workflow orchestration engines (Temporal, Airflow, or similar).
• Hands-on Docker: authoring Dockerfiles and running multi-service stacks locally.
• Production AWS or cloud infrastructure experience, including deploying, operating, and debugging cloud services.
• Strong database fundamentals, especially SQL: schema design, query design and optimization, indexing, migrations, transactions, and ORM tradeoffs.
• Real experience with LLM-powered features: agentic workflows, prompt design, and evaluation.
• Fluency with AI coding tools (Claude Code, Codex, Cursor, Superset, Devin) as a daily driver.
• Ability to reason about code and understand what’s under the hood beyond model output.
• Strong communication and product instincts for a collaborative environment.

Nice-to-have

• Experience running multiple AI coding agents in parallel with git worktrees.
• Electron / desktop app development.
• Model routers and multi-model orchestration.
• Self-hosted or open-weight models.
• AWS Bedrock.
• Python.
• FedRAMP, GovCloud, or other security frameworks.
• Multi-tenant SaaS, IaC, or Kubernetes.
• Vector search / embeddings.
• Compliance, supply-chain, or other document-heavy domains.