// EXPERTISE
One senior developer. Backend through frontend.
Everything below is a productized service with a defined scope and a deliverable you can verify. I build the software, and implement performance, technical SEO and analytics in code as part of that build rather than as a separate project afterwards. Pick a service, or tell me what you're building and I'll tell you which one fits.
What's included?
- Node.js, TypeScript, React and Next.js architecture matched to the application and its data flows.
- Responsive interfaces, APIs and backend services delivered as one production-ready application.
- Performance, accessibility, analytics and technical SEO fundamentals considered during development.
- Core Web Vitals budgets enforced in the build: LCP ≤ 2.5s, CLS = 0 targets from the first commit.
- E-commerce ready: cart, checkout, payment gateway integration (N-Genius, Adyen, Bank of Muscat), order events.
- GA4 + GTM + Meta Pixel data layer implemented and validated before launch.
Deliverable
A production-deployed website with a Lighthouse report, crawl report and tracking validation sheet.
What's included?
- Node.js services and APIs for business logic, integrations and data-heavy applications.
- Clear service boundaries, validation, error handling and production observability.
- Queues, webhooks, payment flows and third-party integrations built for reliable operation.
- MongoDB, PostgreSQL or MySQL data modelling, migrations and indexing for the queries the application actually runs.
- Health checks, structured logging and error tracking, so a failure surfaces before a customer reports it.
Deliverable
A deployed backend with API documentation, tests on the critical paths and a runbook for operating it.
What's included?
- Next.js App Router applications with server and client rendering chosen per route.
- React interfaces that are fast, accessible and structured for real production use.
- Routing, metadata, structured data, analytics and performance built into the application.
- Rendering chosen per route — static, server-rendered or incrementally revalidated — and written down so it stays that way.
- Core Web Vitals budgets enforced in the build: LCP ≤ 2.5s, INP ≤ 200ms, CLS = 0.
Deliverable
A production-deployed Next.js application with a Lighthouse report and a written record of the rendering decision behind each route.
What's included?
- Legacy platform assessment, URL inventory and migration planning before implementation.
- Redirects, metadata, content, integrations and tracking verified through the move.
- Staged cutover and post-launch checks designed to make the migration measurable and reversible.
- Launch-day QA: crawl comparison old versus new, indexation watch and a rollback plan.
- 30 days of post-launch monitoring — rankings, impressions and CrUX — with weekly written reports.
Deliverable
A migrated platform with a documented redirect map, launch QA record and 30-day monitoring report.
What's included?
- Custom storefronts, booking flows, business applications and backend services.
- Payments, inventory, webhooks, customer flows and integrations where the business requires custom logic.
- Technical SEO, analytics and performance handled as part of the software rather than as a separate speciality.
- Payment gateways integrated and shipped in production: N-Genius, Adyen, Bank of Muscat and PointsPay.
- Product and Offer schema, plus revenue tracking that reconciles against the order table rather than the analytics dashboard.
Deliverable
A live storefront with payment and order flows verified end to end, and a tracking layer that reconciles against your order data.
// WHY ONE ENGINEER
What's the difference between a consultant, an agency, and me?
01
A separate specialist
Audits and recommends. Hands you a document of things your developers must somehow implement. Cannot touch the code. The gap between the report and the result is where the work dies.
02
Typical agency
Has engineers, but in a different department to the person you talk to. Your specification passes through three account managers. You pay for the hand-offs, the overhead, and the meetings about meetings.
03
Working with me
One senior engineer who writes the specification and the code. The integration design and the integration. The redirect map and the migration. Nothing lost in translation, because there is no translation.
// ENGAGEMENT
How do engagements usually start?
Audit
A scoped review of an existing codebase or platform, with the quick fixes implemented rather than listed. Start here when what to build is still open.
Project
A build, migration or rebuild with a written scope, architectural peer review and verification sign-off.
Ongoing support
Continuing engineering on a platform already running: dependency and security updates, monitoring, and a queue of small changes.
Not sure which one fits? Tell me what you're building and I'll tell you.Get in touch
// FAQ
What should you know before picking a service?
If you're unsure, tell me what you're building or what is going wrong. Send a URL or a few lines about it and I'll reply with an honest recommendation, even when the recommendation is that you don't need me right now. No obligation, no sales pitch.
Yes. Most builds are one scope covering the application, the performance budget it has to hold and the instrumentation that reads it, because the same person writes the requirements and the code. Nothing has to survive a handover between the party that specified the work and the party that implements it.
Yes, and that is most of my experience rather than an exception. At Thrifty I worked in a team of five under the Head of E-commerce, alongside a DevOps engineer, a UI/UX designer and two other developers, and set and reviewed the work of an outside development agency of three to five people. Code review, a shared definition of done and someone else owning the roadmap are the normal conditions, not a constraint to design around.
Permanent employment first. I am looking for a senior or principal individual contributor role, working inside cross-functional agile sprints with ownership of real production service boundaries, and that is what this site is primarily for. Project work is still possible and is written down in full before anything starts. No rates are published here, because the same service can be a two week piece of work or a two month one depending on scope.
It leaves artifacts somebody else can re-run: tests on the critical paths, a Lighthouse and CrUX report, a redirect test matrix for a migration, an event validation sheet where tracking is in scope, and a runbook for operating what was built. Every one of them is reproducible by another engineer, which is the point. You never have to take my word for it.
// OPEN TO WORK