Job description

ABOUT COFOUNDER At General Intelligence, we’re building highly-capable autonomous agents for startups. Our goal is to enable the one-person, one-billion-dollar company. Our product, Cofounder, enables entrepreneurs to start and run companies. Using it should feel more like a company-building game where users coordinate agents, unlock roadmap nodes, run missions, build artifacts, grow departments, and make revenue go up. WHAT YOU’LL OWN - End-to-end product work across the agent experience: flows, integrations, and UI/UX - Flows: creation, editing, run criteria, execution reliability, and observability - Integrations: add new integrations and deepen existing ones - UI/UX: work on novel, interesting ways of building human<>agent interactions - Product sense for complex UX: making powerful systems feel simple and reliable - Ship features across the stack with a tight feedback loop from real users (frontend + backend) - Improve performance, reliability, and cost across agent runs and flow execution - Build instrumentation, analytics, and developer tooling to speed up iteration - Maintain a high bar for code quality, tests, observability, and documentation REQUIREMENTS - 4+ years building and shipping user-facing products end to end - Frontend: React/Next.js (or equivalent), TypeScript; component-driven development, performance profiling, accessibility, data fetching/state, optimistic UI and cache invalidation - Backend: Python FastAPI, background workers/queues, webhooks, Postgres (or similar) schema design, transactions, migrations - Solid testing and observability habits (unit/integration tests, metrics, traces, structured logs) - Bias to ship: scope thin slices, deliver quickly, instrument, iterate NICE TO HAVE - Event-driven architectures, WebSockets/real-time UIs - LLM/agent experience: tool-use patterns, prompt hygiene, guardrails, evaluation - Infra familiarity (Vercel/AWS/Modal/Supabase/Redis Cloud or equivalent), CI/CD, feature flags WHY JOIN - Ship directly to engaged users; your work is visible and high-impact - Build the core of an autonomous agent platform with real integrations and real workloads - Small, elite team; high ownership; fast decisions; zero bureaucracy

Requirements

Must-haves

End-to-end product work across the agent experience: flows, integrations, and UI/UX
Flows: creation, editing, run criteria, execution reliability, and observability
Integrations: add new integrations and deepen existing ones
UI/UX: work on novel, interesting ways of building human<>agent interactions
Product sense for complex UX: making powerful systems feel simple and reliable
Ship features across the stack with a tight feedback loop from real users (frontend + backend)
Improve performance, reliability, and cost across agent runs and flow execution
Build instrumentation, analytics, and developer tooling to speed up iteration
Maintain a high bar for code quality, tests, observability, and documentation
4+ years building and shipping user-facing products end to end
Frontend: React/Next.js (or equivalent), TypeScript; component-driven development, performance profiling, accessibility, data fetching/state, optimistic UI and cache invalidation
Backend: Python FastAPI, background workers/queues, webhooks, Postgres (or similar) schema design, transactions, migrations
Solid testing and observability habits (unit/integration tests, metrics, traces, structured logs)
Bias to ship: scope thin slices, deliver quickly, instrument, iterate
Ship directly to engaged users; your work is visible and high-impact
Build the core of an autonomous agent platform with real integrations and real workloads
Small, elite team; high ownership; fast decisions; zero bureaucracy

Nice-to-haves

Event-driven architectures, WebSockets/real-time UIs
LLM/agent experience: tool-use patterns, prompt hygiene, guardrails, evaluation
Infra familiarity (Vercel/AWS/Modal/Supabase/Redis Cloud or equivalent), CI/CD, feature flags

Application prep

Use the role’s language and application details while tailoring your materials.

Keywords to mirror

goreactnext.jstypescriptaccessibilitypythonpostgresllmawsci/cd

Platform: ashby
Effort: quick