Agentic AI & RAG Architecture Review with Soham Kshirsagar

Bring your use case or current design for an agent or RAG system. We cover retrieval quality, chunking, orchestration, tool design, evaluation and cost, and you leave with concrete next steps.

If you're building RAG (Retrieval-Augmented Generation) systems but struggling with architecture, retrieval quality, or scaling - this session will help you.

We will discuss:

• Designing production-grade RAG architectures

• Choosing the right vector database (Pinecone / Weaviate / FAISS / Chroma)

• Improving retrieval quality and context relevance

• Handling hallucinations in RAG systems

• Chunking strategies and embedding optimization

• Structuring LLM pipelines for real-world applications

We can also review your existing RAG project and discuss improvements.

Best suited for:

• Developers building GenAI applications

• Engineers working on document AI / knowledge assistants

• Backend engineers integrating LLMs into products

• AI developers facing retrieval or hallucination issues