Skip to content
The record

The work, in two collections.

Not summaries — full engineering narratives. Production Systems prove senior ownership inside a team; Independent Projects prove the same muscles solo, from zero. Each opens with the problem that forced the decision.

Production Systemsteam · Adit platform
01

Booking Availability Engine

Flagship

A public scheduling API from twelve seconds to under one — the fix was in the architecture, not the algorithm.

Read the story →
02

Kafka Event Platform

Why exactly-once delivery wasn't the answer — idempotency was. 14.6M+ messages.

Read the story →
03

EHR Integration Framework

~60 systems you don't control, mapped onto one model — and the abstraction I refused to build until the cases forced it.

Read the story →
04

Bidirectional Synchronization

A dirty flag beat a consensus protocol — the overwrite race, solved with a boolean.

Read the story →
05

Production Operations

The work that starts after merge — the staged release path and the live-ops that keeps thousands of clinics running.

Read the story →
Independent Projectssolo · product → deploy

SprintDay

live

A twenty-module product built solo — then the productionization tax: migration-drift recovery and a 929MB → 391MB image.

Read the story →

JSONit

PWA · offline

A client-side JSON toolbox whose value is its performance architecture — the parsed document never leaves the Web Worker.

Read the story →

A self-hosted platform (Docker · Caddy · per-project DB isolation · encrypted off-box backups) runs the independent projects — and more experiments are in progress.