Selected work

The through-line is still trust: strong systems, clear evidence, fewer brittle handoffs.

The projects span AI products, modernization programs, and cloud-backed applications, but the goal is consistent: build something other people can rely on after the launch energy wears off.

01

AI-enabled product workflows

RAG-informed systems, retrieval tuning, and production workflows that prioritize grounded outputs and safe behavior.

Representative proof

TwinMail - AI-native email workflows

Offline .mbox ingestion supports archives up to about 55 GB.

02

Operations modernization

Workflow redesign and automation for high-friction operational environments using SharePoint, Power Platform, and structured process controls.

Representative proof

EPIC operations modernization

Paper-heavy workflows were shifted into self-service systems.

03

Cloud and data foundations

Pragmatic full-stack architecture with API services, PostgreSQL plus pgvector, and deployment patterns focused on maintainability.

Representative proof

TwinVault - secure AI-enhanced asset platform

Role-aware controls and audit-friendly lineage were built into the platform from the start.

Trust layer

Proof behind the delivery themes

Trust here comes from published work, verifiable public links, and product surfaces that point to real systems rather than generic portfolio chrome.

The through-line

Case studies, proof bands, and targeted contact paths make it easier to inspect the work without turning the site into a pitch deck.

Next step

Need deeper context?

Ask assistant

Resume is useful for role-shaped evidence. Assistant is useful for fast answers.

Anonymous answers only search public content

3 free left

Assistant

Ask assistant

Use for fast, citation-backed answers from published work.

Use case studies for narrative proof. Use resume view for role-shaped evidence.