writings

i write about what i learn

oct 2025 fixing rag retrieval with semantic double-pass chunking

sep 2025 intelligent context management: system design for ai agents

jul 2025 building a context relevance classifier with llama 3.2

jun 2025 training llama 3.2-3b to think better: grpo-lora-rl

jun 2025 colqwenrag: vision-based rag for legal-ai papers

jun 2025 fine-tuning embedding models for legal-rag

mar 2025 building custom reasoning models with grpo and sft

mar 2025 how to make models "think": the deepseek approach

mar 2025 fine-tuned a llm to draft indian legal contracts