Services

Senior capability, without the layers.

Every engagement is led end to end by the same person who scopes it — from stakeholder conversations and architecture decisions to production .NET and Azure code. Fragmented accountability is one of the quietest killers of technology programmes; we remove it by design.

The record on this page was earned by our founder over 25 years in UK public-sector IT — most recently as a senior consultant at one of the world's largest consultancies, for UK government clients. The wins belong to those teams; the method and the experience are what you hire here.

Legacy modernisation

£7m+
across seven programmes
0
downtime or significant defects

Classic ASP, VB6, .NET Web Forms — systems everyone agrees are a problem, attached to programmes nobody quite believes in. Our founder has modernised that exact estate, repeatedly: seven programmes over four years, all delivered within budget, with no downtime and no significant defects.

The method is UI-driven, combining the strangler fig and anti-corruption layer patterns — engineered to keep the risk you accept visible, bounded and small. It moves in three deliberate steps.

01 · Prove

Proof before promise

A working demonstration on your actual system within weeks — not a slide deck of projected benefits.

02 · Run side by side

Old and new together

Both systems serve real users from day one — no cutover weekend, no leap of faith.

03 · Replace & accelerate

Incremental, AI-assisted delivery

Functionality moves across in deliverable slices, with guardrailed AI agents making the legacy archaeology dramatically faster.

Legacy system Classic ASP · VB6 · Web Forms still running — the safety net Modern replacement .NET · Blazor · Azure built slice by slice ANTI-CORRUPTION LAYER One database — your trusted data
Old and new operate on the same data. No big bang, no migration cliff — every step reversible.

Target platform: modern .NET, Blazor and Azure, aligned to the GOV.UK/GDS Service Standard where it applies. It's been done before — seven times — by the person who'd be doing yours.

DELIVERY OVER TIME → 01 MODERN 02 MODERN 03 MODERN 04 MODERN 05 LEGACY 06 LEGACY 07 LEGACY the line moves right, one shipped slice at a time
Each slice ships visible value on its own — better screens, faster tasks — while risk stays bounded.

Greenfield applications

15+ years
our flagship build, still in production
~£150k/yr
saved against the product it replaced

New line-of-business systems, designed and built to last. The document imaging system our founder conceived and built — a solution nobody had even asked for — is still in production 15+ years later with no significant defects, saving around £150k a year. That longevity isn't luck; it comes from holding the vision of what's being built while staying hands-on with the code.

We move from a budget-holder's boardroom, to a whiteboard wireframe, to production .NET and Azure code without dropping the thread — so what gets built is the system the business actually needs, not the one the brief described.

Agentic AI delivery

~40%
raw productivity uplift, with quality improving
9 months
ahead of the mainstream, in live production delivery

Our founder was running structured human-and-AI delivery around nine months before "agentic software engineering" had a name — the first on a major government account to bring AI agents into a live production delivery workflow.

We'll be direct about what that experience showed: the gains are real, but they are not automatic. An AI agent pointed at a legacy codebase without structure will hallucinate confidently and waste your team's time. The same agent, embedded in a disciplined workflow — right context, right guardrails, engineers who verify its output — lifted raw productivity around 40% while improving quality.

We stand that model up inside your team: guardrails, Git safety nets, reviewable commits, and the delivery patterns that make AI output trustworthy enough to act on.

Engineers verify and direct AI agents accelerate the work Guardrails patterns · context · Git safety nets Shipped reviewed & verified ~40% productivity uplift — with quality improving, not eroding.
AI output is only useful when it's trustworthy enough to act on. The guardrails are the product.

Software architecture & delivery

~£30m
hybrid-cloud programme, lead design
10/10
client satisfaction on that programme

Solution architecture and lead design experience on programmes up to ~£30m — including a hybrid-cloud transformation that modernised 30+ legacy applications to Azure and AWS PaaS and scored a perfect client satisfaction rating of 10.

Domain-driven design, message-driven architecture, Azure DevOps CI/CD, and the unglamorous groundwork — authentication, pipelines, hosting design — done before anyone asks for it, because when the opportunity arrives, "give me six months to set up" is the same as "no".

Fractional CTO & delivery rescue

£750k £1.3m
a failing programme rescued, then grown
10
engineers recruited and built into the delivery team

Senior technical leadership without a full-time hire: technology roadmaps that bridge what's possible and what's affordable, team building and mentoring, and honest counsel for budget-holders making high-stakes decisions.

When deliveries slip, we take them over. Our founder drove a £750k government programme in serious difficulty to delivery by financial year-end — then grew it into £1.3m of work that would otherwise have been zero. A £5m+, four-application modernisation was delivered on time, in budget, with zero defects above cosmetic, by a team of 10 he recruited and built.

kick-off beyond the engagement Meladian involvement a safety net, not a bottleneck your team's capability self-sufficient handover
Success is not being needed: patterns, guardrails and mentorship are part of every delivery.
The commercial deal

Whatever the problem, the terms are the same.

We don't do time and materials — that just moves the delivery risk onto you. Every Meladian engagement works the same way:

A fixed price for an agreed outcome

Like a builder quoting a house. If it costs more than we said, that's our problem, not yours.

Stages of six weeks or less

Each stage ends in working software with stand-alone value — and a genuine exit point.

You own everything we deliver

All IP and code is yours from day one. Emerging dependencies are called out early and designed away — no lock-in.

Free until we both believe in it

The conversation and our first read of the problem cost nothing. You get a fixed quote before any money changes hands.

The full step-by-step is on the home page.

Which of these is your problem?

Start with the free conversation — and if a proof of concept is the right first step, it runs on your actual system: weeks, not months, reversible at every step.

Start the conversation