Systems that make AI trustworthy in production.

Problem

A production GenAI / RAG stack was shipping changes with no reliable way to know whether retrieval quality had regressed for real customers.

Approach

Built a golden dataset grounded in real customer utterances and automated daily regression runs. Led embedding-model evaluation (BGE-M3, Stella 1.5B) for the retrieval path, and shipped per-utterance explainability — MRR, mAP, Hit-Rate on every query — so non-ML stakeholders can debug a bad score without filing a ticket.

Impact

Problem

A mature low-code platform used by tens of millions needed generative & agentic capabilities without breaking the reliability its scale demanded.

Approach

Led generative & agentic AI efforts reimagining Power Apps and Power Pages — embedding AI into the authoring experience while holding the bar on performance and availability for a massive install base. Much of this work remains confidential.

Impact

Problem

Cloud applications needed data protection & privacy as a first-class, reusable capability — not something each team reinvented under deadline.

Approach

Designed and built a data-protection-and-privacy microservice on the cloud platform, packaged so any application could consume it — the architecture that became a granted patent. In the same period, architected cloud-native microservices on Cloud Foundry for multi-cloud deployment.

Impact

Problem

A 30-year-old product with 10K+ live customers had accumulated the drag of three decades — slow, dated, and hard to love.

Approach

Headed the modernization effort and shaped design, strategy and roadmap directly with Senior Directors — treating a legacy renovation as a product problem, not just a rewrite.

Impact