Writing
Reinforcement Learning for Design Taste
Training Qwen3.8-27B on MiMo's webdev RL environments, and how the choice of LLM judge shapes a model's design taste.
Exploring Similarity Search with Prompted Subspaces
Using text prompts to choose which kind of similarity an image search uses.
Training Search Agents with GRPO
Educational blog post for training search agents with GRPO over a financial-filings dataset.
