NLP

Corpus

An open evaluation harness for RAG systems: golden sets, retrieval metrics, LLM-as-judge with calibration, and regression gates in CI.

Why build another one

Because "it seems better" is not a release criterion. Corpus turns every prompt change into a pull request with a diff of measurable outcomes.

Calibrating the judge

An LLM judge is a model, so it needs its own eval. We label 200 examples by hand, measure the judge's agreement with the humans, and only trust it above κ = 0.7.

More systems

AI Agents

Atlas

An agentic retrieval system that decomposes a question, routes across five source types and returns an answer with paragraph-level citations.

Python
LangGraph
pgvector
Cohere Rerank

Computer Vision

Sentinel

Real-time defect detection running at 60 FPS on a Jetson Orin Nano — quantised, distilled and deployed to 140 production lines.

PyTorch
TensorRT
ONNX
OpenCV

LLM Apps

Ledger

Structured extraction from 40-page financial PDFs with schema validation and a confidence score per field, replacing a nine-person review team.

Python
Pydantic
Claude
Tesseract