Jarmin
Member of Technical Staff (Models)
San Francisco, CA, US · Staff · Full-time · Posted Aug 16, 2026
- Experience
- 3+ years of experience
- Visa sponsorship
- US citizen or visa holder only
About the job
Own the post-training loop to improve AI Employees by defining and running model training and evaluation work.
What you'll do
- Post-train and fine-tune models, including turning agent traces and expert feedback into training data.
- Develop reward models and graders for non-verifiable outcomes, and distill models into smaller, faster, cheaper models.
- Design and run training experiments.
What they ask for
- Have trained or fine-tuned models and shipped the result into a product.
- Have hands-on experience with SFT, preference optimization, or RLHF.
- Have strong software engineering fundamentals alongside ML depth.
- Want to work in person in San Francisco.
Nice to have
- Have adapted open-weight models to a specialized domain or built training data or evaluation infrastructure.
- Contribute to open source projects.
Tools
SFT, preference optimization, RLHF
AI focus
Improve AI Employees through post-training, fine-tuning, reward modeling, training-data development, and model distillation.
Model-written summary · the original posting is authoritative