About the role
The team works with AI companies to refine open-source models for specialized tasks. The researcher finds creative ways to use data to build training pipelines and environments that improve models. These models are then deployed to reach many users.
Our summary of the posting; the original is on the Baseten careers page.
What the posting requires
- Required
- SFT
- Training environments
- Customer
- Training experiments
- Publish
- Reward engineering
- GRPO
- Evals
- Production data
- Reward hacking
- Preferred
- RL training frameworks
- NeurIPS
- ICML
- ICLR
- LLMs
- Reward modeling
These tags are our reading of the posting. They can miss something; the original posting is the reference.