Show HN: Shrewd – what I learned distilling LLM labels into local classifiers
shrewd Turn LLM judgments into a small, fast, local text model for one fixed task. I built shrewd while making local models for another project. It started with a question: can GEPA prompt optimization get better labels from an LLM?
- ▪shrewd Turn LLM judgments into a small, fast, local text model for one fixed task.
- ▪I built shrewd while making local models for another project.
- ▪It started with a question: can GEPA prompt optimization get better labels from an LLM?
Hacker News (AI / LLM) files mainly under ai. We currently carry 6,053 of its stories.
Story provenance
Source · retrieval · rights · ranking — open for full record
inspect →
Story provenance
Attribution is not the same as permission. This drawer separates discovery metadata, excerpts, WeSearch-generated summaries, reuse status, and whether the publisher receives the visit. Nothing here claims a legal grant the publisher has not made.
Record
| Original publisher | GitHub |
| Canonical URL | https://github.com/sshah03/shrewd |
| Publication time | Tue, 22 Sep 2026 20:00:46 +0000 |
| Retrieval time | 2026-09-22T20:03:57.143Z |
| Last seen | 2026-09-22T20:03:57.143Z |
| Headline source | Publisher (no WeSearch rewrite) |
| Excerpt source | publisher body |
| Excerpt method | First ~120 words (~800 chars) of extracted publisher body, fair-use limited. |
| Summary | WeSearch · cerebras-chat (WeSearch summarizer) |
| Summary source text | contentText |
| Citation coverage | Summary is a WeSearch-generated derivative; primary citation is the original publisher URL. |
| Cluster | GOWjq0B134Pv · 1 stories |
| Cluster logic | Grouped by semantic title/content similarity across sources within a rolling window. Same-publisher template collisions are excluded from coverage comparison. |
| Ranking reason | Story pages are not engagement-ranked. Hub feeds use recency, with optional source-diversified chronological ordering (cap consecutive stories per source). No personalized ranking. |
| Publisher visit | Yes — open original |
| Substitutes article? | No — link-out required for full text |
Rights status (four layers)
WeSearch handling by dimension
| Indexing | May the item be indexed (stored, ranked, made findable)? | Allowed |
| Snippet | May a short excerpt of the publisher's text be shown? | Allowed |
| AI summary | May WeSearch generate its own short summary of the article? | Limited |
| Retrieval / RAG | May the content be exposed for third-party retrieval-augmented generation? | Not asserted |
| Model training | May the content be used to train AI models? | Not asserted |
| Commercial reuse | May the content be reused commercially? | Not permitted |
Basis: Derived from the published RSS/Atom feed. Contact: [email protected]. Reviewed: 2026-07-24.
Opening excerpt (first ~120 words) tap to expand
shrewd Turn LLM judgments into a small, fast, local text model for one fixed task. I built shrewd while making local models for another project. It started with a question: can GEPA prompt optimization get better labels from an LLM? Then, since Jev-style decisions have gotten popular recently, I wanted to see how far I could get toward them locally for a fixed set of questions. This repo contains the pipeline I used and what I measured along the way. It is useful when you repeatedly classify similar text, have a few hundred hand-labeled examples and a larger unlabeled pool, and care about inference cost, latency, or keeping text on your own hardware after training.
…
Excerpt limited to ~120 words for fair-use compliance. The full article is at GitHub.