(Remote) Founding Full-Stack Engineer

Role Specifications

The construction industry loses $1.6 trillion annually to inefficiencies — not because people are careless, but because critical knowledge is trapped in PDFs, transcripts, and scattered systems. Every new project starts from scratch. We're building the opposite: an AI co-worker that extends the Projektsteuerer (the person who runs complex construction projects), while every project makes our system measurably better. The moat isn't the model — it's the compounding project memory no competitor can replicate. Pre-seed led by Realyze Ventures — LPs include Zech and other major European construction groups. Co-investors: D11Z (the family office behind Aleph Alpha) and the CDTM Venture Fund (backed by 300+ CDTM alumni including founders of Personio, Alasco and the Technical Director of DeepMind). Our software is running today on a major autobahn construction program and an S-Bahn transit program — multi-year timelines, hundreds of thousands of pages of specs, protocols, and communications. Real consequences when we get it wrong. Tasks Two hard problems. All production. AI-native project management, not a chatbot on top of PDFs The construction industry doesn't need another chat wrapper around GPT. It needs project management software where AI runs in the background — through cronjobs, proactive alerts, and task dashboards — not in a sidebar chat window. The hard product question is how AI shows up: when as silent background automation, when as a surfaced recommendation the PM approves, when as a fully delegated task the user only reviews after the fact. Extending harness engineering all the way to the UI Agent harnesses don't stop at the backend. A multi-step agent run that streams tokens, calls tools, hits failure modes, and recovers gracefully needs a frontend that makes all of that legible in real time. You'll plumb Vercel AI SDK streams from our agent layer straight into the component tree — streaming, tool-call traces, interrupts, partial renders, graceful failure — all as first-class UI concerns, not afterthoughts. Stack: TypeScript, Next.js, Vercel, Supabase (Postgres + pgvector), LangChain, Vercel AI SDK, LangFuse, shadcn/ui. €500/month AI tooling budget per engineer — Claude Code, Cursor background agents, experimentation with frontier models. No legacy. Greenfield. Requirements Who we're looking for Product sense: you reason from user pain → solution → measurable outcome. You can talk to non-technical customers and understand their workflows. Velocity + craft: prototype fast, measure everything, iterate on real feedback. But you care about reliability because the downside of wrong is real in construction. Comfort with the unknown: many of these problems don't have Stack Overflow answers. You read papers, prototype, and compare approaches. Bonus: strong open-source work, previous early-employee startup experience, domain depth in document understanding or agent systems, or shipped systems that replaced hours of human labor. Lev

Stack & Technologies