I helped build Adobe Commerce. Agencies call me when the answer isn’t obvious.
At Adobe I shipped features into the product itself — RMA, Staging, Analytics,
Commerce Cloud, the Migration Tool. Those commits are still public.
Thirteen years since then inside other people’s builds: the architecture decision that
takes a year to undo, the bug that has already beaten three engineers, and the honest read
on whether a project is worth signing at all.
Years on Magento
13
Projects delivered
100+
Clients worked with
40+
Peak store load
150k/day
Support status
2.4.xSupportedPatched — exact window depends on the minor version
2.3.xEnd of lifeSecurity patches ended September 2022
2.2.xEnd of lifeSecurity patches ended December 2019
2.1.xEnd of lifeSecurity patches ended June 2019
2.0.xEnd of lifeSecurity patches ended March 2018
1.xEnd of lifeSecurity patches ended June 2020
Reading which row a store sits on is the easy part. The expensive question is what the
move does to your integrations, your customisations and your delivery schedule — and
whether the answer is an upgrade at all or a rebuild.
That call is what I get brought in to make.
What I get called for
Decisions that are expensive to reverse
Multi-store and multi-source modelling, B2B structure, headless against Luma, Edge Delivery, integration topology. Get these wrong and the project pays for it for two years.
Technical read before you sign
Is this build winnable, at what price, and where is the risk actually sitting? I ran this call in pre-sales across 40+ deals — including the ones worth walking away from.
The escalation nobody can close
The L3 layer, after your team has ruled out the obvious. Usually somewhere between a plugin, a cron and an assumption everyone inherited.
A build that is slipping
When the delivery date is at risk and the thing actually on the line is the agency's relationship with its client, not the ticket.
Migrations, in both directions
Onto Adobe Commerce from anything, and off it to anything. I have no reason to keep you on a platform that has stopped fitting — and I will say so.
Overflow you can put your name on
White-label senior capacity — no headcount, no recruitment, and I never appear in front of your client.
Track record
2024–now
Independent — agencies & merchants
Upgrades, performance audits and white-label engineering, contracted directly or behind an agency's brand.
2021–2024
Lead Magento 2 engineer — commerce agency
Owned delivery on the agency's heaviest stores and ran the technical side of pre-sales — the audits that decide whether a deal is worth signing.
150k daily visitors
40+ clients, EU & US
Team 7 → 15
2019–2021
Tech lead — SaaS product company
Designed an eCommerce management platform from the architecture up, then built the delivery process around it: CI/CD, code review, release management.
17+ releases shipped
CI/CD from zero
Team built from scratch
2018–2019
Magento 2 core work — independent
Stayed in the Magento 2 codebase after Adobe, on core-level fixes and module work.
2016–2018
Core engineer — Magento Inc. / Adobe Systems
On the product itself, not on projects built with it. Shipped features into the commercial release and signed off architecture inside an Adobe engineering organisation.
127 commits in core
12+ features shipped
RMA · Staging · Analytics
Commerce Cloud
Migration Tool
2014–2016
Senior engineer — London agency
Out-staffed into other companies' teams on Magento EE and Cloud — the same white-label model I run independently now, learned from inside.
7+ companies
12 projects delivered
Team 5 → 16
Credentials
Commits in the Magento 2 core repositoryPublic sourceVerified
127 commits · Dec 2016 → Jul 2017magento/magento2Check ↗
UTC+2/+3 — full overlap with UK and EU, US Eastern to mid-afternoon
White-label
Standard. I stay behind your brand and off your client calls unless invited
Starting point
A straight written answer on whatever you are weighing, before either side commits
Platforms
Adobe Commerce, from the inside
I worked on the product itself at Adobe, so I read the codebase as something I helped write rather than something I reverse-engineer. B2B, Enterprise and Cloud, Magento 1 CE and EE, and Edge Delivery Services.
Architecture, upgrades, performance under real traffic, and the escalations that have already defeated the team.
B2BEnterpriseCommerce CloudMagento 1 CE / EEEdge Delivery Services
Shopify
Theme and app development, Plus builds, and the integration layer that connects a store to the ERP, tax engine and payment stack behind it.
Most often I am called in when a Shopify build has outgrown what apps can do and needs real engineering underneath.
ThemesAppsPlusIntegrations
WordPress
WooCommerce stores and content platforms, including headless setups where WordPress runs the content and a separate commerce engine runs the transactions.
Useful when the catalogue is small but the content and SEO surface is large.
WooCommerceHeadlessContent platforms
Migrations, in both directions
Onto Adobe Commerce from any platform, and off it to any platform. Data model, integrations, order history and SEO intact — the parts that quietly break six weeks after launch.
I have no reason to keep you on a platform that has stopped fitting. If the honest answer is that you should leave Magento, that is the answer you will get.
Commerce data does not stay in the store. Catalogue feeds in, order and customer data out to ERP, tax, analytics and whatever the finance team actually uses — and all of it has to keep moving when volume spikes.
I build that layer: queue-driven ingestion and ETL pipelines on RabbitMQ and AWS SQS, warehouse and search indexes that stay in sync, and AI-assisted ingestion using OpenAI APIs and n8n. Running on AWS, Google Cloud or Azure, with the observability to know when a feed has silently stopped.
Retained maintenance and escalation cover across all three platforms — not just project work. Patch cycles, monitoring, and someone senior on the other end when production misbehaves at 9pm.
Agencies use this as bench cover: I am there when the client escalates and invisible the rest of the month.
Retained coverPatch cyclesEscalationMonitoring
Infrastructure
AWSEC2 · RDS · S3 · SQS · CloudFront
Google CloudCompute · storage · managed data
AzureCompute · storage · pipelines
The platform is half the job. I also own what runs underneath it — and I have moved
stores between clouds often enough to say which parts of a migration are routine and which
ones cost you a weekend of downtime.
Why go direct instead of through an agency or a body shop?
Because you get the engineer, not a margin. When you buy senior Magento work through an intermediary, a large share of the rate pays for the layer in between, and the person who actually opens the codebase is whoever was free that week — often not the one from the pitch deck.
Direct, you get one named person who stays on the account, and the whole rate buys engineering. You also get an answer instead of a ticket. No account manager relaying a technical question to someone you cannot speak to, and no three-day round trip to find out whether something is possible.
The trade is real and worth stating: an agency gives you a bench and holiday cover. I am one person. That is why I work best as senior capacity alongside your team, not as a replacement for it.
How do I know the work is actually senior?
Check it rather than take my word. The Adobe certifications are verifiable at Adobe with a credential ID, and the 127 commits in the public magento/magento2 repository carry my name in the git history — that cannot be edited or claimed by anyone else.
Beyond that, ask me something hard on the first call. Ten minutes on your actual architecture will tell you more than any CV.
Do you work white-label under our brand?
By default, yes. I use your repositories, your process and your tracker, I stay off client calls unless you invite me, and I never approach your client. As far as they are concerned the work came from your team, because it did.
What does an engagement cost?
Day rate or fixed scope, agreed before anything starts. I quote after I understand the problem, not before — a number given blind is either padded for risk or wrong.
Short diagnostic work is often the cheapest way to find out whether a bigger engagement is even worth doing. The first written assessment costs nothing.
How fast can you start, and how do you communicate?
Usually within two weeks, sometimes sooner for a contained piece of work. Full-time or part-time.
I work UTC+2/+3, which is a full overlap with UK and EU hours and covers US Eastern into mid-afternoon. Whatever your team already uses — Slack, Teams, Jira, Linear — I work inside it rather than asking you to adopt anything.
What happens if you are unavailable mid-project?
Fair question and the honest weak point of hiring one person instead of a firm. I manage it by writing things down: decisions, architecture notes and runbooks live in your repository, not in my head, so your team can carry on without me.
For work where continuity is critical, say so at the start and we scope it so nothing depends on a single person being online.
Who owns the code and the IP?
You do, on delivery. Standard contractor terms, work-for-hire, and I sign your NDA and your agreement rather than pushing my own paperwork.
Would you tell us to move off Magento?
If that is the right answer, yes. I have no platform to defend and no licence to resell, so recommending a move to Shopify or a rebuild costs me nothing and saves you a year.
An advisor whose income depends on one answer is not an advisor.
Get in touch
Need someone senior on the team? Write, call, or just ping me — whichever is easiest.
A build you need owned, an architecture decision you are weighing, an escalation your
team cannot close, ongoing cover, or a second opinion on someone else’s design.
If you would rather start small, send me the question you are stuck on — you will
get a straight written answer, free, and you owe nothing either way.