AI-Native Product Engineer for Desktop AI App

We’re looking for an experienced, product-minded engineer to help us test, debug, and iterate on Prism, an internal AI desktop application that we have built and want to finalize. Prism is already in active use and working, but it is a complex product with many moving parts: chat-based AI asstitance, agents, knowledge/workspace context, artifacts, productivity workflows, and an Electron desktop shell. We need someone who can help us make it more reliable, polished, and usable. The ideal person thinks like a product manager and can develop. We want someone who can use the product, find rough edges, reproduce bugs, write clear issues/specs, fix TypeScript/React/Electron problems, and add or improve tests where needed. Helpful experience: - AI-native products, chat interfaces, agents, LLM workflows, or tool-calling systems (Stack includes Mastra, Assistanz UI) - Electron, React, TypeScript, Node, pnpm/monorepos - Playwright, Vitest, E2E testing, screenshot-based QA, regression testing - Spec-driven or acceptance-criteria-driven development - Debugging complex app state across frontend, backend, IPC, persistence, and async workflows - Product sense: cares whether the user experience actually works, not only whether tests pass Typical work may include: - Dogfooding Prism and finding UX/reliability issues - Reproducing bugs with clear steps and evidence - Turning product gaps into short specs or acceptance criteria - Fixing bugs across renderer/main-process code - Iterating with us on chat, agents, artifacts, and core workflows This is a complex, ambitious product that already works and is used internally. We’re looking for someone pragmatic, AI-native, detail-oriented, and comfortable helping turn a working internal product into something robust enough for wider use.