@mohitvverma
AI/ML Engineer | Building production RAG, agents, and LLM systems
Collaborating on AI/ML projects, technical writing, discussing production ML architecture
1d ago · 2 min read · Stop Using Fixed-Length Chunking: The 1 Change That Gave Us 40% Better RAG Precision The Problem: 6 Months of Optimization, Silently Undermined Our production RAG system handles 12K daily queries. We're a team of 4 ML engineers who spent 6 months tun...
Join discussion1d ago · 2 min read · GraphRAG Beats Vector Search by 86% — But 92% of Teams Are Building It Wrong GraphRAG beats flat vector search by 86% on multi-hop queries — but most teams are building it wrong. They bolt Neo4j onto LangChain and call it done. Missing: Entity resol...
Join discussion2d ago · 14 min read · Stop Using Fixed-Length Chunking: The 1 Change That Gave Us 40% Better RAG Precision We spent 6 months optimizing embeddings, HNSW params, and prompts — then swapped chunking strategy in 2 hours and beat everything. Here's the embarrassing truth. Fou...
Join discussion2d ago · 10 min read · Stop Using Fixed-Length Chunking: The 1 Change That Gave Us 40% Better RAG Precision We spent 6 months optimizing embeddings, HNSW params, and prompts — then swapped chunking strategy in 2 hours and beat everything. Here's the embarrassing truth. Fou...
Join discussion
2d ago · 7 min read · GraphRAG Beats Vector Search by 86% — But 92% of Teams Are Building It Wrong Microsoft's GraphRAG paper showed that graph-structured retrieval with community summarization significantly outperforms flat vector search on multi-hop and thematic queries...
Join discussion