A record of work I can talk about publicly. Long-form write-ups are being published progressively — entries marked coming soon outline what each will cover. Several significant engagements aren’t represented here at all because the details are sensitive; I’m happy to discuss those directly under appropriate agreement.
A fully automated AI system that calls and texts patients to check on their wellbeing, scores every conversation in real time for clinical risk, and hands off patients-in-need to live staff — operating under HIPAA and clinical-safety constraints from day one.
A production system that reads ~3 million inbound clinical documents monthly — ER records, specialist notes, labs, faxes — and uses vector embeddings to surface high-risk conditions for clinical review in near real-time.
A purpose-built change data capture engine replacing commercial CDC tooling across five heterogeneous source systems — 8 million events per day at baseline, peaks over 40 million, end-to-end latency consistently under two minutes.
Near-real-time CDC capture from PostgreSQL into Snowflake, plus high-volume file-processing infrastructure, for a company that manages hospital patient flow, capacity, and throughput — HIPAA-bound and operationally critical.
Moving 24 Oracle instances onto a 13-site multi-master MariaDB topology that had to survive county-level network outages and keep elections running. Delivered three months ahead of a $12M contract. Still at 100% uptime through hundreds of election cycles.
An orchestration abstraction layer managing the full server lifecycle across 110,000 servers, 20 data centers, and seven regions spanning public cloud, FedRAMP, and private cloud — replacing an operational estate of 60,000 hand-run scripts.
The internal migration runbook for Amazon's company-wide mandate to move applications off VMs and onto AWS — written before CloudFormation, Lambda, ECS, or most managed services existed, when everything was built from EC2, S3, EBS, and early RDS.
Financial-grade MySQL architecture in the pre-synchronous-replication era: multi-master replication for trading accounts and positions, heavy sharding, 7,000+ QPS sustained — architected from outside, then operated from inside.
A 90-day rolling HTTP log warehouse built on MySQL 5 with MERGE-table partitioning, MEMORY-engine ingestion staging, and IOPS-aware hardware design — solving the data-warehouse problem years before the tools that now define it.
A full multi-tenant platform for the direct-sales industry — CRM, replicated distributor websites, marketing automation at 22M+ messages daily, compensation-plan management, payments, ecommerce — built from primitives before vertical SaaS had reference architectures.
Chef-based infrastructure-as-code for MySQL at cloud scale — dynamic tuning, replication topology management, automated rebuilds and backups — that became the foundation of HP's commercial Database-as-a-Service offering.
Migrating an AI-driven supply-chain forecasting platform from MariaDB onto SingleStore, where a single HTAP engine handling both transactional and analytical workloads fit the shape of ML-driven operations better than separate systems.
A multi-year database consulting relationship at one of the largest credit unions in the United States: MySQL cluster builds, PostgreSQL platform engineering, and Patroni-based high-availability design under NCUA regulatory constraints.
Under appropriate agreement: Fortune 100 database architecture for MariaDB enterprise accounts, additional public-sector engagements, and the engagement details that don’t belong on a public page. Get in touch if any of this is relevant to work you’re doing.