Founding Engineer

We are looking for a founding engineer who wants to build deeply technical AI systems in one of the most important and regulated industries in the world. You will work directly with the founders to: Build agentic workflows for clinical trial automation Design reliable systems for document ingestion, data transformation, validation, and auditability Own core product surfaces across frontend, backend, data infrastructure, and AI orchestration Work closely with clinical teams to turn messy real-world workflows into production software Ship quickly while maintaining the rigor required for regulated healthcare environments You might be a fit if you: Are an exceptional full-stack or backend engineer Care about building reliable, production-grade AI systems and can/have integrated into legacy systems in the past Move fast, take ownership, and like ambiguous technical problems Are excited by using AI to automate healthcare, clinical trials, or regulated data workflows Want to join at the earliest stage and help define the engineering culture We care less about credentials and more about raw ability, taste, speed, and ownership. This is a high-agency founding role for someone who wants to build category-defining software. Astraea is building the AI operating system for clinical trials. Our product helps clinical teams automate the work between a study protocol and FDA-ready outputs, including protocol and SAP ingestion, clinical data review, SDTM and ADaM dataset generation, TFLs, QC, traceability, and audit trails. On the technical side, Astraea combines AI agents with deterministic data processing. The agent layer helps interpret study documents, plan workflows, and coordinate tasks, while the actual clinical outputs are produced through structured, validated Python and R pipelines. This gives us the flexibility of LLMs without relying on free-form model output for regulated clinical data work. The desktop product is built with React, TypeScript, and Electron, with embedded Python and R runtimes for local clinical data processing. We use schema validation, audit logs, provenance tracking, versioned outputs, and human-in-the-loop review to make the system reliable and reviewable. Send us any relevant information (resume, background, github) Saffron interview 30 min technical with Sanmay (CTO) and Eng Live coding session with Eng team 30 min culture interview Paid work trial on a realistic Astraea engineering problem. This is designed to evaluate how you think, build, communicate, and handle ambiguity. Can accomodate online if needed. Offer