A patented AI system that predicts food expiry and quality loss from shelf metadata — cutting waste before it happens.
Retail loses a significant share of perishables to spoilage. Expiry labels are conservative estimates, so stores over-stock and discard food that is still perfectly safe.
Predicting true shelf life from a single static label ignores storage conditions, batch variance, and handling — so waste stays both an operating cost and a sustainability problem.
Built a feature pipeline around product metadata, batch history, and storage signals.
Trained gradient-boosted models to predict per-batch quality decay instead of a single expiry date.
Ranked stock by predicted freshness so replenishment and markdowns happen before spoilage.
Packaged the models behind a small REST API with clear prediction contracts.
The model doesn't predict an expiry date — it predicts a decay curve. That turns a conservative label into a per-batch, per-condition decision.
1patent filed
food-tech system for freshness prediction from shelf metadata
RAG-grounded LLM workflows that answer enterprise questions from real company knowledge — not a demo corpus.
100% grounded — every answer traced to a retrieved source — never generated from memory
The REST APIs and workflows behind Peoplebox's core HR modules — one service layer powering six product areas.
6 product areas — Goals, Reviews, Surveys, HRIS, IDPs, and reporting on one service layer
EXPLAIN-first schema and query work that kept enterprise dashboards and workflows fast as data grew.
p95 response time — flattened on million-row tables by index-backed, EXPLAIN-first query plans