← Founder Notes
Archive

Code search just learned to answer why. entire's agentic search, out september 4, blends semantic…

Yethikrishna ROriginal on Threads

code search just learned to answer why. entire's agentic search, out september 4, blends semantic search across commits and session transcripts so agents see the reasoning behind a change.

the codebase got a memory.

Context

Entire's blog post, listed September 3, 2026, says it is releasing Agentic Search, one API across every repo an agent can access, returning the code, the reasoning and the session transcript behind a change. It combines semantic search over session transcripts and commits with exact code search, stores chunks in turbopuffer, and has regional deployments in the US, EU, India and Australia. It works through the Entire CLI with major coding agents.

How it compares

The post is listed September 3 and the note says September 4, a one day gap that may be timezone. Entire's benchmark of nine real engineering history questions, ten runs each, has a baseline agent with Git history and the GitHub CLI answering 70 of 90 correctly and the agent with search 81 of 90, using less than half the tokens and steps. It is vendor-reported, the sample is small and the question set is the vendor's, so it is not independent. The codebase got a memory is the author's take.

Watch next

  • An independent evaluation and per-repo availability.

Sources

  1. Entire: introducing Agentic Search for code and contextentire.io

Provenance

The note above is reproduced unedited from the original post, first published on Threads on 22 September 2026 at 23:17 IST. Sources are the papers and datasets the note draws on.

View the original post
Embed this note
<iframe src="https://founder.myndlabs.tech/notes/embed/code-search-just-learned-to-answer-why-entire-DdmTkHqja58" width="480" height="420" style="border:0;max-width:100%" loading="lazy" title="Code search just learned to answer why. entire's agentic search, out september 4, blends semantic…"></iframe>

More notes