Experience
The most consequential technology decisions rarely look like technology decisions in the moment. They look like budget conversations, vendor negotiations, hiring choices, and architectural debates. The job has always been to see them for what they are — and govern them accordingly.
The Journey
Engineering was always the direction. The draw was toward making things work — not just code, but systems, products, teams, and the thinking underneath them. Built from first principles. Learned by doing.
At Samsung, led a task force in unfamiliar territory — smart device development and cross-device communication — with no existing playbook and a tight brief. The approach was deliberate: divide the problem, assign deep ownership across the team, and synthesise. The expertise the team built didn't stop at the brief. It was taken into a global Samsung hackathon in Seoul, where the team placed first in its category.
As a founder, the safety net disappears. Every technology decision carries a capital consequence, a risk attached, a tradeoff to defend. That weight changes how you think. At some point the question shifted from 'can we build this?' to 'should we — and at what cost to the organisation?' Everything since has followed from that shift.
Consequential Decisions
Migrated a live, client-facing application with zero downtime — no documentation, a reluctant outgoing team, and real reputation risk on the line. The migration was completed within 3 days. A full rollback strategy was built and tested before touching production. The outgoing team recommended downtime and a like-for-like move; the harder, better path was chosen — and the application came out structurally stronger.
Designed and deployed an enterprise-grade multi-component solution, then deliberately chose not to activate approximately 70% of the planned infrastructure at launch. Capital was provisioned for scale. It was not spent until scale arrived — protecting the client's runway without compromising the architecture.
Paused an entire project mid-delivery when a client's shifting requirements had already caused more than a month of delays, eroding timeline, budget, and team coherence. Sat down, restructured the engagement, formalised change requests going forward, and recovered the incurred loss before resuming. A difficult conversation that protected everyone involved.
Advised on mobile DevOps for an application with over 100,000 users that was losing ratings due to slow, manual release cycles. Redesigned the pipeline end-to-end — automated builds, structured approval gates, Slack-integrated alerts. A release cycle that once took a week was cut by over 80%. Ratings recovered. Feedback loops tightened.
On a fast-moving first-mover product, kept weekly team velocity deliberately capped and staged the release so the user-facing experience shipped first — polished and testable. Real users were onboarded within 2 months of launch while the admin interface was built incrementally behind it. Speed with structure, not at the expense of it.
Beyond work, reading widely, writing occasionally, and tending a small garden with more patience than urgency. Time with family — including two kids who ask surprisingly sharp questions — keeps the thinking grounded and honest.