SlipstreamJobs tracks this role from the company's public career site. Apply directly on the employer's site.
TypeSafe AI is an AI lab building production-ready AI systems that combine LLM reasoning with code reliability. The company recently launched Jev, their first System One Model, and is backed by a16z with $870M in Series A funding.
You will build data-tooling interfaces, datasets, and evaluation frameworks that directly improve model intelligence and production reliability. This is a full-stack role combining product sense, data science, and engineering. You'll own domains end-to-end: creating high-leverage datasets and UIs that unlock new capabilities, building internal evaluation and debugging interfaces, developing frameworks to measure model quality and reliability, running rigorous experiments to understand how data and training choices impact behavior, and translating real-world customer needs into concrete model improvements.
The tech stack is primarily Python, with TypeScript, Next.js, and Tailwind CSS for frontend, and Kubernetes for orchestration. You'll work closely with product managers and customers, iterating rapidly in an ambiguous space to discover new techniques for building trustworthy AI products.
TypeSafe is a small, flat, close-knit team working fully in-person from their San Francisco office near Embarcadero station. The culture emphasizes craftsmanship, emotional honesty, and collaboration.
REQUIREMENTS:
- Generalist with solid fundamentals and strong coding ability
- Product intuition and high attention to detail
- Hands-on experience implementing LLMs; deep understanding of their capabilities and limitations
- Scrappy, hacky mindset; thrive in ambiguous problem spaces
- Ability to think beyond code to real-world system usage
- Critical perspective on LLM outputs; don't trust blindly
