← All work
Live · Work

Ask Bin Anything

A RAG agent over my public writing — ask it things when I'm not around.

Try:
Tap a preset above or type your own question below.
>

Answers are generated from my public material and may be wrong. For anything important, contact me directly.

How it works

Diagram: a question flows left-to-right through vector retrieval and a language model, emerging as an answer with numbered citations.QuestionVector retrievalLLMAnswer [n]content/*.mdx
  1. 01

    Your question becomes a vector

    Your question is matched against public bio, project, experience, writing, and publication excerpts.

  2. 02

    Top snippets pack into a prompt

    The four most relevant chunks are assembled with instructions to answer in your language and cite sources inline as [n].

  3. 03

    A concise answer streams back

    A non-reasoning language model replies in 2–3 sentences with inline citations you can click to verify.

Under the hood

Next.js 15 with content indexed into SQLite as embeddings at build time. Per-IP rate limit (10/hour), site-wide cap (100/day), 24-hour answer cache. An automatic provider-fallback chain keeps the widget answering even when the primary model is rate-limited or down.

Honest limits

It answers from the public material indexed on this site. Dates matter, and it can miss or misunderstand details; check the cited sources. For unpublished work or personal decisions, contact me directly.