SlipstreamJobs tracks this role from the company's public career site. Apply directly on the employer's site.
Pelico is a manufacturing orchestration platform that uses AI to connect fragmented supply chain, production, and quality data for industrial operations. The company serves major manufacturers including Airbus, Boeing, Safran, Eaton, and Cartier, and has raised $72M from General Catalyst, 83North, and Serena with 300% YoY revenue growth.
You would be a founding engineer on Pelico's new agentic platform (Condor), working on a small, high-bar team building AI agent capabilities for real manufacturing customers. This is a hands-on, full-stack role shipping real features end-to-end, not demos.
Key responsibilities:
- Ship agent capabilities end-to-end: build features from tool/MCP and orchestration backend through React console, live for users
- Harden the runtime: implement durable execution, retries, scheduling, and human-in-the-loop approvals for agent actions with real consequences
- Make agents trustworthy: grow the eval harness (scenario evals, LLM-as-judge) to catch regressions in agent behavior before shipping
- Build a reliable console: real-time streaming of agent events, tool calls, and approvals that controllers can trust
- Set craft standards: write specs others build against, maintain frontend/backend contracts, establish reusable patterns
You'll benefit from Pelico's existing assets (data, event dependency graphs, simulation engine, next-best-action algorithms) and access to real factory data from multiple leading manufacturers.
Qualifications:
- Excellent full-stack engineer who ships: strong code quality with real experience building across Python backend and React frontend (or clear ability to get there fast)
- Hungry and driven: want a hard problem to own, not a comfortable seat; have the ceiling to grow significantly
- Rigorous with non-deterministic systems: spec and eval before shipping, trust evidence over intuition, own what ships including AI-assisted code, follow problems into production
- AI-native and curious: understand how agent frameworks, tools, and evals work under the hood; use AI coding tools daily
- Calm in fast-moving ambiguity: stay effective on a stack where best practices change quarterly, turn uncertainty into workable plans
- Nice to have: real AI/agent product experience (MCP, agent frameworks, evals); durable workflow engines (e.g., Temporal); streaming UIs; observability (OpenTelemetry); spec-driven development; multi-tenant SaaS; aerospace/defense/discrete-manufacturing context
