Tools & Search
Give your agents the tools to act on the real world.
An agent without tools is just a chatbot. The moment you give an agent the ability to search the web, query a database, or look up information in a knowledge graph, it becomes genuinely useful. This track covers the specialized tools that make production AI agents powerful: Tavily (search built for LLMs), hybrid semantic+keyword search, and Neo4j GraphRAG.
Standard vector search is great for finding semantically similar content, but it misses exact matches. Hybrid search combines dense semantic embeddings with sparse BM25 keyword matching — the best of both worlds. For domains where relationship context matters (medical records, legal documents, organizational charts), knowledge graphs stored in Neo4j can answer questions that pure vector search cannot.
Whether you're building a research agent that searches the web, a document assistant that retrieves from a corporate knowledge base, or a recommendation system with complex entity relationships, this track covers the retrieval architecture you need.
📚 Learning Path
- Tavily LLM-optimized search
- Vector search fundamentals
- Hybrid dense+sparse retrieval
- Neo4j knowledge graphs for RAG
- Tutorial: Visualizing graphs with Neo4j Bloom