Software engineer — backend & platform
Nine years of making services that start fast, fail loudly and read clearly at two in the morning. I work on the layer most people never see — queues, migrations, build pipelines, the deploy that has to go out on a Friday.
A replacement ingestion pipeline for an analytics product — cutting a nightly batch job down to a streaming path that settles in under a minute.
Everything I can find on schema evolution and on why teams abandon their own internal tools after eighteen months.
Two days a week from next quarter, for contract work or a long-running advisory seat with a small platform team.
Selected work
A nightly batch that regularly overran its window, replaced by a streaming path with replayable offsets. The hard part was never throughput — it was migrating eleven downstream consumers without a freeze.
Twelve teams, twelve different ways to ship. I built the boring paved road: one manifest, one CLI, sensible defaults, and an escape hatch for the two teams that genuinely needed one.
A small retail client losing orders to timeouts at the worst possible moment. Idempotency keys, a proper outbox, and an honest retry story. Nothing clever — just finished.
Toolkit
None of this is a religion. Pick the thing your team can debug at 3am, then stop talking about it.
Go
TypeScript
Python
SQL
Postgres
Kafka
Redis
ClickHouse
Kubernetes
Terraform
OpenTelemetry
GitHub Actions
Design docs
Incident review
Mentoring
Deleting code
How I work
Long discovery phases hide risk rather than remove it. I'd rather put a thin version in front of real traffic early and let it tell us what the plan got wrong.
More about my background →Repos, dashboards, the last six incident threads. I want to know what your team already tried before I suggest anything.
A short design note: the problem, two options, the one I'd pick and what it costs. If it can't fit on a page, I don't understand it yet.
Smallest end-to-end path that touches production, behind a flag, with metrics on it from day one. Then widen it.
Runbooks, a walkthrough recording, and two weeks of pairing so the system belongs to your team and not to me.
“He spent the first week reading our code instead of rewriting it. By the time he proposed anything, he knew the system better than most of us did.”
“Our deploys used to be a ceremony with three people watching. Now a junior ships on their second day and nobody holds their breath.”
“The documentation he left behind is the reason we didn't need him after the contract ended. I think that was the point.”
Before we talk
Three-to-six month contracts where there's a specific system to build or rescue, and ongoing advisory days with teams who mostly need a second opinion they trust.
Almost always with a team. The work is worth more when someone internal is pairing on it, because they're the ones carrying the pager afterwards.
A weekly rate for contracts, a day rate for advisory. Both quoted after a call, both fixed for the length of the engagement so nobody is counting hours.
Remote by default, with overlap on European hours. I'll come in person for kickoff and for anything that genuinely needs a whiteboard and a room.
Pixel-level front-end design, and rewrites proposed before anyone has measured the thing they want rewritten. I'll usually argue for the smaller fix first.
Taking enquiries for next quarter
A paragraph is enough to start. If it isn't something I should take on, I'll say so and point you at someone better placed.
Typical reply
Within two working days
Working hours
09:00 – 18:00 GMT
Also happy to
Review a design doc, free