December 2025 Summaries

Exa has launched a state-of-the-art people search benchmark to enhance its AI search engine capabilities, particularly benefiting sales, recruiting, and market research teams. The new system involves a hybrid retrieval method combining Exa embeddings with Exa search, capable of indexing over a billion people and handling 50 million updates weekly. The initiative includes the release of an open-source people search evaluation, featuring a dataset with 1,400 queries and an evaluation harness for reproducibility. Analyzing 10,000 historical people queries revealed three primary search categories: role-based search, skill or role-based discovery, and specific individual lookup. Data generation for these categories involved creating synthetic queries for executives and founders across various company sizes, ensuring profiles have public verification but are not widely known to prevent memorization. The evaluation methodology employs recall and NDCG metrics for targeted lookups and uses an LLM judge for role-based discovery due to the subjective nature of results, validating matches through online profile content.

Exa 2.1 is a significant update to the Exa search engine, offering improvements in search quality and performance for both fast and agentic search needs. The update enhances all Exa search API endpoints, including Exa Fast, Exa Auto, and Exa Deep, with the Exa MCP quality benefiting from a new deep search tool. Exa Fast now provides highly accurate search results in under 500 milliseconds, outperforming typical search APIs that rely on Google and cannot achieve such low latency. For users prioritizing quality over speed, Exa Deep delivers the highest accuracy through agentic searches. The company has developed its own document index and retrieval algorithms independently, allowing for these advancements. Evaluation methodologies include low-latency and MCP tests, with results benchmarked against expected answers using advanced grading systems. The development reflects years of research and innovation in semantic and lexical databases, with plans for further scaling and the upcoming release of Exa 2.2.