OctenAI (@OctenAI) header
OctenAI (@OctenAI) profile picture

Search infrastructure for AI. LLM-native real-time web search, extract & embeddings for agents. The complete retrieval stack.

San Francisco, CA
Joined June 2026
Octen Pulse, our new search API, returns results in 10ms, fast enough for models to search while they reason. It's open source, with a tech report on how we built it. Try it and tell us what you build. Link below.
1
5
10
210