// 🚀 Just shipped a production-grade RAG system using LangChain + Pinecone + GPT-4! The retrieval accuracy improved by 40% using hybrid search (BM25 + semantic).
142 28
Multi-Agent AI1 week ago
// 🤖 Multi-agent workflows are the future of AI automation. I built a research agent pipeline using CrewAI that autonomously scrapes & summarises papers.
89 15
MLOps2 weeks ago
// 💡 Hot take: most ML models fail in production not because of poor accuracy, but poor engineering. Observability is non-negotiable.
CONNECTED.
LinkedIn
@vinaysiddha
GitHub
@vinaysiddha
Twitter
@vinaysiddha
// 🚀 Just shipped a production-grade RAG system using LangChain + Pinecone + GPT-4! The retrieval accuracy improved by 40% using hybrid search (BM25 + semantic).
// 🤖 Multi-agent workflows are the future of AI automation. I built a research agent pipeline using CrewAI that autonomously scrapes & summarises papers.
// 💡 Hot take: most ML models fail in production not because of poor accuracy, but poor engineering. Observability is non-negotiable.