Exa

Semantic search engine that finds relevant content (not just keywords) for AI agents.

Overview

Exa uses neural search to find websites and documents that are semantically similar to a query, rather than matching keywords. It is specifically optimized for LLM agents, returning high-quality, ranked results that reduce hallucination in RAG pipelines.

Best for

• Semantic RAG retrieval
• Finding specific types of documents (e.g., PDFs, GitHub repos)

Trade-offs

What XeroHack pre-wires

When you pick Exa in the interview, the engine emits these files into your scaffold:

• Exa Search
• Exa Find Similar