Work
Five case studies, one template: the problem, what I built, what changed, and what’s under the hood.
Q.Promo AI
RetailRetail analysts spent days answering promotion questions. This system answers most of them in about ten seconds.
Quail
HealthcareNHS trusts receive thousands of patient complaints a year, each read and typed up by hand. This platform does the reading.
AI Ticket Routing
Real EstateCurrently building an AI agent that resolves customer tickets using a knowledge base — and knows exactly when to hand off to a person instead.
Product Attribute Accelerator
RetailE-commerce teams typed product attributes in by hand, one product at a time. This system reads the description and the photos and fills them in.
Class Action Early-Warning System
AutomotiveBuilt for an automotive client in 2020: a system that read public complaints and flagged the defect patterns likely to become class-action lawsuits.
Earlier work
Q.Shelf
Assortment algorithms for retail range and space decisions at Quantium.
Transaction Enrichment
Transaction enrichment for global banking clients at Quantium.
Refinery Process Optimization
Predictive maintenance modelling for oil & gas refinery operations at LatentView.
Field Service Benchmarking
Field service performance benchmarking analytics at LatentView.
Anomaly Detection
Anomaly detection across operational data streams at LatentView.
Distress Ranking
Computer-vision distress ranking for aviation infrastructure at TCS.
Credit Risk Engine
Loan-risk machine learning for banking clients at TCS.
FI Transaction System
Banking transaction systems for a global financial institution at TCS.