UK · open source · software craftsperson
mani neomatrix369
Lead software, data & ML engineer building in public — RAG/LLMs, NLP, and production-grade systems. This site hosts live demos and benchmarks; GitHub holds repos, katas, and curated learning lists.
Live on this site
- Model extraction playground Latest + 20 historic snapshots · 111 scored runs · rankings & heatmaps
-
Doubleword extraction guide
Video walkthrough ·
extractor.py --all-doubleword - UK Charity Doc Extract Benchmark write-up · provider findings & methodology
- py-bug-trace sweep explorer 29 models · Laguna-XS.2 rank 12/29 · heatmap & Pareto
- py-bug-trace · Laguna eval One-pager, write-up, technical report & level scorecards
- Tripwire dashboard Security & quality scanning prototype
On GitHub
- awesome-ai-ml-dl Curated AI/ML/DL resources · 1.7k★
- awesome-graal GraalVM & Truffle resource list
- RagCheck Proactive RAG corpus quality assessment
- laguna-eval-experiments Poolside Laguna hackathon · py-bug-trace source on Hugging Face
- playgroup_202602_docextract Source repo for the doc-extract benchmark
Open doc extract benchmark All projects GitHub profile @theNeomatrix369
Browse demos & pages
Pick a card below — interactive demos open in the browser; static pages are project summaries and guides. Cards are grouped by project family (tag on the left).