Document AI · Our own product · 2021–2026

TalkingDB document question-answering

A question-answering engine that rebuilds each customer document as its section tree and fetches only the sections a question needs.

What I did

Wrote the first working version of the tree-based answering engine
Removed leaked credentials from repository history instead of relying on key rotation
Held back an unproven accuracy claim before it reached customers

2026 · TalkingDB document question-answering

Situation: Reviewing repositories before a host move, I found OAuth client secrets in git history, which ships to anyone who clones.
Task: Make the secrets unreachable, not merely invalid, before the repositories were pushed to their new home.
Action: I cloned bare mirrors, mapped each leaked value to a placeholder, rewrote history with filter-repo, pushed to the new remotes, and documented the exact procedure.
Result: The repositories moved with secrets removed from every commit, and another file with hardcoded credentials was flagged for cleanup.

2026 · TalkingDB document question-answering

Situation: A large reported jump in search accuracy was circulating and was about to become a go-to-market claim.
Task: Check whether the gain was real, since repeated passing results seemed to be counted again across runs.
Action: I raised it plainly, then set a general standard: proposed improvements need a frequency table of failing cases and expected fix share.
Result: The inflated figure never reached a customer-facing claim, and benchmark results are now reviewed against a sheet before approval.

2026 · TalkingDB document question-answering

Situation: Leadership wanted a cross-product integration delivered sooner because it strengthened the story with channel partners, which I agreed with.
Task: Show that the date was set by available engineering time, not technical complexity.
Action: I agreed with the goal, offered to make time visible through tracked tasks, and named three ways to buy time: delegation, clear ownership, hiring.
Result: The conversation moved from when it can be done to which trade-offs we accept, with the task breakdown as shared evidence.

2026 · TalkingDB document question-answering

Situation: Search training was to be ported from an older e-commerce AI codebase, and the files looked like a simple lift and shift.
Task: Scope the port properly before committing a sequence of stories others would plan around.
Action: I read the code itself: the crawler was tied to one storefront, so I called for a fresh generic crawl-and-chunk pipeline and resequenced.
Result: The order of work changed without changing the plan, and client-specific code stayed out of the generic package.

Write a line; I’ll find it in my inbox.