Skip to content
§ 03 — AI Lab

Experiments.

Fig 0. Field Notes

AI products I’m building — some live, some in the lab. Real problems, real constraints.

Deployed · 1 unit
01Nutrition / Vision
Live

Prato

Snap a photo of any meal and get an instant nutritional breakdown. Uses vision models to identify ingredients and estimate macros — no manual logging, no barcode scanning.

Vision LLMTypeScriptNode.js
Launch →
DEPLOYED
Lab Notebook · 2 units in progress
02Retrieval-Augmented Generation
In Progress

RAG Knowledge Engine

A retrieval-augmented generation system that lets you query any document or knowledge base in plain language. Built to demonstrate production-grade AI data handling — ingestion, chunking, vector search, and grounded LLM responses.

WIP
03Code Analysis / LLM
In Progress

AI Code Reviewer

A developer tool that analyses a GitHub PR or code snippet and returns a structured review — bugs, performance issues, security flags, and improvement suggestions. Built to demonstrate tool calling, structured LLM output, and API integration.

WIP
§ 03.B — Engage

Want to build something like this?
Open a channel.

AI prototypes, infrastructure for AI products, or production systems behind them. Bring the problem; I’ll bring the spec.

Open a channel →