rduffyuk/engineering-memory-benchmark — GitHub trending stats & insights

Featured

rduffyuk/engineering-memory-benchmark

Empirical study: layered retrieval (typed→semantic→grep) scores 0.954 for LLM-generated engineering artifacts. 5 conditions, 3 model tiers, 36 generated ADRs, 23 score files.

Visit GitHub

Python

2 contributors

Custom license

Social mentions

Recent discussions about this repository across the web

Agree partially my layered approach is multi-step self-correction. Typed retrieval surfaces facts, grep catches the errors, output improves across steps.

@rduffyuk · x.com

Results across 5 production ADRs (rubric-scored, dual-judge validated): No memory: 0.572 Semantic search: 0.720 Grep + file read: 0.918 Typed-fact retrieval: 0.650 All three layered: 0.954 Each layer…

@rduffyuk · x.com

No trending activity

This repository has not yet been featured on GitHub Trending

Repository activities

repository's daily and monthly activities across stars, forks, merged PRs, issues, and closed issues