Case studies
Four systems, in depth
Context, scale, what I decided, and what came of it.
01● Current
DolaSoft · an energy company (client under NDA)
An AI-first SDK for ~30 engineers
AI product & developer tooling
Migrating a monolithic industrial control application into a modular SDK that both engineers and LLM agents can build against.
In daily use by ~30 engineers
A frontend team of 4–6 — and I still write code daily
02● Current
DolaSoft · an AI code-review GitHub App
AI code review, zero to production
Agentic systems in production
A GitHub App that reviews every pull request in the context of the whole repository — with an eval harness gating every prompt and model change.
A single AWS Lambda — no database, no framework, no LangChain
Claude on Amazon Bedrock, native tool-use loop
03● 2019–2022
Avallain · one of the world's largest education publishers (client under NDA)
A learning platform used by 1.8M learners
Legacy rescue · technical leadership
Joined as a frontend developer, grew into the technical lead, and rebuilt a legacy platform that the company had already written off as unprofitable.
1.8M learners
High-load delivery on a fixed release schedule
04● 2016–2018
Luxoft · one of Australia's largest investment banks
A deal platform carrying $200M+ of daily flow
Leadership at scale
Deal-processing platform under strict security and compliance, delivered by 6 teams across 4 time zones.
$200M+ daily
6 teams, 30 engineers and QA across 4 time zones