Stephen Donahue
c6d7c8c211
Some checks failed
Deploy to S3 / deploy (push) Failing after 1m24s
Retract the seven published columns into drafts/ and promote the 24 per-date entries split out of Notes/random.md. Bodies are verbatim; titles, subjects and slugs are new. Make publication a matter of location. Notes/Notes.11tydata.js gives anything under Notes/ permalink: false unless tagged `column`; drafts/drafts.11tydata.js renders drafts at their real permalink under `yarn dev` and excludes them from collections entirely in a build, so no template can surface one. Publishing is now `git mv drafts/<file> columns/`, and moving it back retracts it. Untagged notes were previously rendered into _site/ and deployed — including Notes/random.md, which was live at /Notes/random/.
1.7 KiB
1.7 KiB
title, date, layout, tags, permalink
| title | date | layout | tags | permalink | ||
|---|---|---|---|---|---|---|
| Panel of Judges | 2026-07-16 | column.njk |
|
/columns/2026/panel-of-judges/ |
Been a month - wild, unverified ideas
-
How could quantum computing impact LLMs, or other probablistic, non-deterministic models?
-
Can we put type safety, reliability features around LLM invocations? Can we formalize it, as its own kind of process: sequential or concurrent invocations of asynchronous processes.
We do that already with RPCs, web requests, etc.
what are some safety patterns around async await?
- timeouts
- can we explicitly prompt the LLM: You have 500ms to respond or we move on?
- Like an ad network bid: "winning" response wins
- How are they compared? Panel of judges
- At first I thought you could single "identical" judges in front of each submission
- that's counting on each to be trustworthy
- Then I thought about having a panel of judges, quorum style
- Then I thought about a "ranking funnel" a shrinking pool, maybe bracket style (is that a merge-sort elimination?)
- or you grade pools of responses as a collection, instead of individually (can't tell if this would be a fidelity loss or win - would that matter?)
- At first I thought you could single "identical" judges in front of each submission
- this sounds really expensive: time and resources
- 1/N success rate, and what does it take to eliminate the other N-1?
- it sounds much less efficient than developing / evolving a single model session (some kind of hybrid) as long as it survives (and maybe each session can determine that for itself).
- LLM (lives) are not precious. They can off themselves.
- exception handling // what's exceptional when prompting an LLM?
- validations / sanitation