AI systems that prove what they know. Not guess. From knowledge engines grounded in number theory to production platforms across healthcare, commerce, and research.
Flagship product Chefrson — tonight's dinner from what's already in your kitchen. chefrson.com →
SUM Equities builds production software across AI systems, agent observability, knowledge engines, and consumer applications. The portfolio includes a verifiable bidirectional knowledge engine grounded in number theory (SUM), a standards-native agent observability backend (HFAO), and a cross-platform kitchen management system in closed beta on Android with the web surface live.
The work ranges from low-level systems programming (Rust, Zig/WASM) through full-stack web platforms (TypeScript, React, Next.js) to AI/ML pipelines (DSPy, multi-model orchestration, vector search). Everything ships to production — deployed primarily on Cloudflare's edge network, Supabase, and Vercel.
The common thread: taking complex technical ideas and turning them into working systems that people actually use. Not research papers — software.
Verifiable surface
32 claims across 10 projects on the proof page — each tagged by epistemic status and a cited source. GitHub counts refreshed at build time.
Beta
Eat what you already bought: pantry-aware meal suggestions, recipe capture from any source, and nutrition tracking that logs itself when you cook. Web live, Android closed beta.
Beta ★ 9 · last commit 1 month ago · 91 commits / 30d
Cryptographic provenance for AI knowledge transforms. Every transform — render, extract, compose, slider — emits a signed receipt anyone can verify offline. Six-regime compliance validators (EU AI Act, GDPR, HIPAA, SOC 2, ISO 27001, PCI DSS) and a layered sum verify --explain output landed in v0.7.0; current release is v0.8.1. Same bytes verify identically in Python, Node, and modern browsers.
Production
A digital archive exploring human potential and breakthroughs across all ages, proving that brilliance is unbound by time.
4 shipped products · 13 experiments in the Lab