SID

You'll **train models with GRPO** and own the entire training pipeline at an **AI research lab for retrieval**. Your work will directly advance frontier methods in search and reasoning.

You'll join an AI research lab to **post-train reasoning into LLMs** and **design RL training environments for retrieval**. You'll run experiments on frontier models, working on **next-generation vision-first embeddings** in a small, high-intensity startup team.