Backend & Distributed Systems
Systems designed for failure, built for scale, and instrumented so you can actually see what they're doing at 3am.
// technology leader · principal engineer · cto · ai
Backend systems, edge computing and network-layer code — the invisible infrastructure that lets the world's biggest brands bring themselves to market. Written mostly in Go, increasingly alongside AI. Shipped by teams I'm proud to have led.
I've spent my entire professional life as an engineer — as a software engineer, a principal engineer, a CTO and an advisor to founders. The common thread is complexity: the platforms nobody sees, handling the traffic everybody depends on.
Most recently that has meant building extremely complex edge solutions — code that runs as close to the user as physics allows, making decisions in microseconds across a global footprint — and bringing AI into the way engineering teams actually work: integrating it into existing workflows, developing the usage patterns that hold up, and getting the best from it.
I care as much about how a team ships as what it ships. Good architecture and good leadership are the same discipline: reduce coupling, make the right thing the easy thing, and get out of the way.
A few of the shapes the problems have taken.
Designed and built a programmable edge layer executing customer logic in hundreds of points of presence — request routing, transformation and policy at the network edge, with sub-millisecond overhead budgets.
Led the engineering behind platforms that let household-name brands launch, sell and scale online — high-concurrency backends, resilient integrations and peak-day traffic that would flatten most systems.
Low-level code where the internet actually happens: custom proxies, TLS termination, connection management and traffic-shaping — written for throughput, tuned for tail latency.
Integrated AI into existing engineering and business workflows without breaking what already worked — defining the usage patterns, guardrails and feedback loops that turn a novelty into a dependable multiplier for teams.
Fractional and full-time CTO roles, plus advisory seats with early-stage founders — turning ideas into working prototypes, prototypes into products, and first hires into engineering organisations.
Systems designed for failure, built for scale, and instrumented so you can actually see what they're doing at 3am.
Moving compute to where the user is. Global footprints, tight latency budgets, and the hard consistency trade-offs that come with them.
Protocols, proxies, sockets and the unglamorous plumbing that decides whether a platform is fast or merely functional.
My language of choice for a decade of production systems — simple, boring in the best way, and brutally fast when treated with respect.
Building and running engineering organisations: hiring well, setting technical direction, and creating the conditions for teams to do their best work.
Helping founders and engineers navigate the messy middle — from first prototype to first million requests.
Integrating AI into existing workflows, developing the usage patterns that actually stick, and getting the best from it — practical leverage, not hype. The goal is never "use AI"; it's shipping better, faster, with fewer surprises.
“The best infrastructure is invisible. You only notice it when it's gone.”
I've spoken at some of the industry's biggest stages and mentored founders through the earliest, hardest stages of building.
Booking a conference or podcast? I'd love to hear about it.
Advisory, leadership roles, speaking, or a genuinely hard engineering problem that needs a second pair of eyes — my inbox is open.
[email protected]