---
format: "aidr-story-markdown/v1"
id: "3af3dbfde98de28b427535216bf33f9b791cc6e34204a1ee9b1c21d52f21be55"
canonical_url: "https://aidr.today/3af3dbfd?lang=en"
title: "LLM Classification Is Feature Engineering"
lang: "en"
requested_lang: "en"
available_langs: ["en","vi"]
translation_fallback: null
fallback_fields: []
published_at: "2026-09-17T15:40:17.000Z"
category: "Research"
topics: ["llm","feature-engineering","classification","machine-learning"]
source_urls: ["https://minimallysufficient.com/posts/llm-classification-is-feature-extraction/","https://news.ycombinator.com/item?id=49742437"]
summary: "LLMs make good classifiers but even better features"
---

# LLM Classification Is Feature Engineering

> [Open the canonical story](<https://aidr.today/3af3dbfd?lang=en>)

**Published:** 2026-09-17T15:40:17.000Z
**Category:** Research
**Topics:** llm, feature\-engineering, classification, machine\-learning

## Summary

LLMs make good classifiers but even better features

## Sources

- [Story source](<https://minimallysufficient.com/posts/llm-classification-is-feature-extraction/>)
- [Discussion](<https://news.ycombinator.com/item?id=49742437>)