A seed-stage AI startup is hiring a Founding AI Engineer to join a mission-driven team applying agentic AI to the built world. The company focuses on multimodal agents that support architecture and building-code workflows, with the goal of helping apartments get built faster.
What you will be doing:
Generate hypotheses and run time-boxed experiments to improve the accuracy of multimodal AI agents.
Reflect on experiment results and update mental models to direct your own research roadmap.
Work on supervised fine-tuning, model distillation, auto-research, retrieval tasks, and context engineering.
Run computer vision evals and build internal tooling to accelerate experimentation.
Talk to users and dogfood the product weekly.
This is an on-site, high-autonomy role for someone who blends software engineering speed with ML rigor and wants to help address the housing crisis at an early-stage company.
- 2+ years in software engineering or ML, with focus on AI/LLM applications
- Highly autonomous and self-directed: able to manage your own workload and generate experiments without being told what to do
- Daily usage of AI coding tools (e.g., Claude Code, Cursor, or similar)
- Blends SWE speed with ML rigor: not too researchy, not lacking experimentation skills
- Experience with Python, multimodal LLMs, computer vision, AI agents, and evals frameworks
- Must work in-person 5 days/week in San Francisco
- Resonates with the mission of addressing the housing crisis and wants to work at a seed-stage startup