AI Engineer
Summary
Founding-team AI engineer at an early-stage company building a foundational model from scratch (novel architecture, not fine-tuning). Day to day: model training, reinforcement learning, and architecture work, plus shipping production LLM systems (agents, RAG, serving) and evals/benchmarks. Core stack: Python and PyTorch, on-site in San Francisco.
About the Role
Join the founding engineering team at an early-stage AI company developing a foundational model from scratch with a novel architecture, rather than fine-tuning existing models. This hands-on role contributes to core model development as well as practical, production-ready LLM systems.
What You'll Do
Build and improve foundational models across training, reinforcement learning, and model architecture.
Develop and ship production LLM applications, including agents, retrieval-augmented generation pipelines, or model-serving systems.
Create evaluations, benchmarks, and quality monitoring for LLM systems.
Contribute as a hands-on individual engineer across model development and application work.
What We're Looking For
One to five years of professional engineering experience.
Practical experience building foundational models across the model-building pipeline, including training, reinforcement learning, and architecture.
A track record of building and shipping production LLM applications.
Strong Python skills and hands-on experience with an ML framework such as PyTorch.
Experience evaluating LLM systems through evaluations, benchmarks, or quality monitoring.
Interest in a ground-level individual contributor role.
Compensation & Benefits
Base salary range: $180,000 to $220,000 USD annually.
Location
On-site in San Francisco, California. Candidates should be based in the San Francisco Bay Area or willing to work on-site in San Francisco.