Legacy application modernisationRebuilds the in-house
team can run.

Strangler-pattern migrations off the system nobody wants to touch, one bounded piece at a time, with the old and new running side by side until the numbers match.

Book a consult See our work

One seam at a time

No big-bang cutover.

Parity measured

Old vs new, same inputs.

Knowledge captured

Docs where your team looks.

Engagement

Built · fixed scope per increment

Each piece moves only when old and new give the same answers. You can stop after any seam and still have a working system.

What you get.

Strangler-pattern migration: one bounded piece at a time, never a big-bang cutover

Old and new running side by side until the numbers match

Parity measured: same inputs, same outputs, in your environment, before anything is switched off

Knowledge captured where your team looks, not in our heads

Engagement: Built · fixed scope per increment · Ask about your migration →

Good for

— A system nobody wants to touch

— A framework three major versions behind

— A migration that has been ‘next quarter’ for two years

We will say no when

— Rewrite it all is the plan, and the only plan

— Nobody can say what the system does today, and nobody wants to find out

Stack · only what we have shipped to production

Monolith → microservices.NET Framework → .NETIIS / EC2 stabilisationRDS migrationsCron → queuesgRPCGraphQLWordPress → custom

From the record.

All case notes →

01 · Digital signage · enterprise

Built · embedded

Signage platform across four continents

Problem — A UK signage business operating across Australia, the US, China and APAC needed a platform its display-manufacturer partners could plug into.

What we did — .NET 6 microservices and Angular on AWS (EKS, ECR, RDS, CloudWatch) with Docker-based CI/CD. Scheduled sync integrations with SharePoint, Power BI and partner sources. Production infrastructure stabilised under sustained memory pressure ahead of a database end-of-support migration.

Measured — Shipped and running across four regions; partner integrations for three major display manufacturers — [ADD UPTIME / DEVICES].

.NET 6AngularAWS EKSRDSDockerCI/CDPower BISharePoint

Client withheld · details on the callAsk about this →

02 · Retail · integrations

Built

Retailer replenishment SDK

Problem — Retailers integrating a replenishment service each failed in their own way.

What we did — .NET Core microservices on Azure Service Fabric with EF Code First, plus an admin module that resolved SDK-level failures across retailer integrations.

Measured — SDK in production across retailer integrations — [ADD RETAILERS].

.NET CoreAzure Service FabricMicroservicesEF Core

Client withheld · details on the callAsk about this →

03 · Enterprise · .NET

Embedded

Process-documentation platform backend

Problem — A browser-extension product that writes process documentation from recorded sessions, with a milestone that could not slip.

What we did — Owned Lambda and hosted worker services (SQS, SNS) across environments and led the refactor that shipped the critical milestone on deadline.

Measured — Critical milestone shipped on the date.

.NETAWS LambdaSQS / SNSWorker services

Client withheld · details on the callAsk about this →

How it runs.

The same loop as every engagement

Week 0

Map the seams

What the system does, who depends on it, and where it can be cut without anyone noticing.

Week 1

First seam, parity test

The smallest valuable piece chosen; a test that replays real inputs through old and new.

Week 6

First seam cut over

New code taking real traffic for one bounded piece; the old path still there, one flag away.

Then

One seam per increment

The same loop, piece by piece, each one measured, each one reversible.

Fortnightly

The review

What moved, what matched, what the old system still owns. Then the next seam.

Questions we get about this.

Longer than a rewrite is promised to take and shorter than a rewrite actually takes. The first seam is live in six weeks; after that you can see the rate and decide the pace.

Then the parity test is the first thing we write: real inputs recorded, replayed through both paths, differences listed. It is the only safe way to learn what the system does.

Yes, and the system is better than when we started: the pieces already moved keep running, and the map of the rest is yours.

The other seven.

All eight end in something running

01Custom software→02Team augmentation→03Data engineering & analytics→04AI & machine learning→05Cloud & DevOps→06Legacy modernisationhere07Mobile→08QA & test automation→

Legacy modernisation · the call

Name the system nobody touches. We will say where to cut first.

Bring the problem, not a brief. If it is outside what we do well, we will say which of our contacts to call instead.

Book a consult See our work

Or email info@hopsig.com — reply within one working day.

What you get.

Strangler-pattern migration: one bounded piece at a time, never a big-bang cutover

Old and new running side by side until the numbers match

Parity measured: same inputs, same outputs, in your environment, before anything is switched off

Knowledge captured where your team looks, not in our heads

Engagement: Built · fixed scope per increment · Ask about your migration →

Legacy application modernisationRebuilds the in-houseteam can run.

What you get.

From the record.

Signage platform across four continents

Retailer replenishment SDK

Process-documentation platform backend

How it runs.

Map the seams

First seam, parity test

First seam cut over

One seam per increment

The review

Questions we get about this.

How long does a migration take?+

What if the old system has no tests?+

Can we stop halfway?+

The other seven.

Name the system nobody touches. We will say where to cut first.

What you get.

From the record.

Signage platform across four continents

Retailer replenishment SDK

Process-documentation platform backend

How it runs.

Map the seams

First seam, parity test

First seam cut over

One seam per increment

The review

Questions we get about this.

How long does a migration take?+

What if the old system has no tests?+

Can we stop halfway?+

The other seven.

Legacy application modernisationRebuilds the in-house
team can run.