WORK
What changed after we left.
One engagement for each capability. What the client faced, what we did, and what their team runs today. Client names are withheld; we can share references on a call.
- engagements shown
- 18
- areas of work
- 4
01 Cloud & infrastructure · AWSUK fintech scale-up The review passed. New environments are now created from code, by the client’s own team.
CHALLENGEOne AWS account held everything, most of the team had admin access, and a customer security review was coming.
WHAT WE DIDWe split workloads into separate accounts under AWS Organizations, moved access to IAM Identity Center with least-privilege roles, and put the whole setup in Terraform.
02 Cloud & infrastructure · Microsoft AzureUK professional services firm One consistent Azure estate, with access rules the IT team can explain and change.
CHALLENGESubscriptions grew without a plan. That was how the client put it on the first call.
WHAT WE DIDWe built a landing zone with separate subscriptions, lined identity up with the existing directory, and moved workloads across in planned waves.
03 Cloud & infrastructure · Google CloudEuropean data analytics company Projects and access are tidy, and cluster upgrades happen on a calendar.
CHALLENGEProjects and permissions have sprawled. That was how the client put it on the first call.
WHAT WE DIDWe restructured projects and permissions, put networking and GKE into Terraform, and set an upgrade schedule with the team.
04 Cloud & infrastructure · KubernetesEuropean SaaS scale-up Upgrades are now a scheduled, routine task the client’s platform team does without us.
CHALLENGESeveral clusters built by different teams, all behind on versions, with upgrades nobody wanted to own.
WHAT WE DIDWe consolidated onto managed clusters, moved every deployment to GitOps with Argo CD, and ran the first two upgrades alongside the client’s engineers.
05 Cloud & infrastructure · Cloud migrationUK logistics SME The move finished without unplanned downtime, and the old servers were switched off.
CHALLENGEThe migration has been on the roadmap for two years. That was how the client put it on the first call.
WHAT WE DIDWe grouped workloads by risk, moved them in waves with a rollback for each, and re-platformed the one legacy app that was blocking the rest.
06 Cloud & infrastructure · Networking & edgeEuropean SaaS scale-up DNS changes are reviewed like code, and staff reach internal tools without a VPN.
CHALLENGEOur network layout can’t grow any further. That was how the client put it on the first call.
WHAT WE DIDWe redesigned the network layout so it could grow, put DNS under version control, and replaced the old VPN with zero-trust access.
07 Platform & delivery · DevOpsUK software company Anyone on the team can release, and a deploy takes minutes.
CHALLENGEDeploys take an afternoon and one specific person. That was how the client put it on the first call.
WHAT WE DIDWe put infrastructure into Terraform, rebuilt the pipelines so they were readable, and made every deploy a single reviewed change.
08 Platform & delivery · Platform engineeringEuropean fintech scale-up New services reach production from a shared template, without a ticket to the ops team.
CHALLENGEEvery team solves deployment its own way. That was how the client put it on the first call.
WHAT WE DIDWe built golden-path templates and self-service infrastructure, starting with the two teams who were waiting longest.
09 Platform & delivery · Software developmentUK media company The service is live, and the client’s own developers now maintain it.
CHALLENGEWe need a service built and don’t have the people. That was how the client put it on the first call.
WHAT WE DIDWe built the service with the client’s product team, took it through load and security testing, and handed over the code with its runbook.
10 Platform & delivery · Databases & data platformsUK e-commerce scale-up Pages stayed fast as data grew, and a database failure is no longer an outage.
CHALLENGEQueries got slow as the data grew. That was how the client put it on the first call.
WHAT WE DIDWe tuned the slowest queries, added a replica with tested failover, and moved reporting off the production database.
11 Security & reliability · SecurityLondon healthtech startup A clear picture of the risks, the serious ones closed, and evidence to show customers.
CHALLENGEIs our cloud actually set up securely? We don’t know. That was how the client put it on the first call.
WHAT WE DIDWe reviewed the cloud accounts against a recognised benchmark, cut admin access back to who needed it, and fixed findings in priority order.
12 Security & reliability · DevSecOpsUK fintech scale-up Security checks run on every change, and developers see the results in their own pull requests.
CHALLENGESecurity review happens at the end, if at all. That was how the client put it on the first call.
WHAT WE DIDWe added dependency, image and secrets scanning to the pipeline, tuned out the noise, and agreed which findings block a release.
13 Security & reliability · Compliance readinessEuropean SaaS company Evidence is now collected automatically and controls are monitored continuously. The client went into its ISO 27001 certification audit prepared, and questionnaires are answered from a trust centre instead of from scratch.
CHALLENGEA large customer asked for ISO 27001 certification. Evidence lived in spreadsheets and screenshots, and every security questionnaire took a week of engineering time.
WHAT WE DIDWe rolled out Vanta, connected it to the cloud accounts, identity provider, code hosting and HR system, then fixed the failing controls it surfaced and wrote the policies to match how the team actually works.
14 Security & reliability · ObservabilityUK marketplace scale-up The team hears about problems from its alerts, not its customers, and the overnight noise has gone.
CHALLENGEWe find out about outages from customers. That was how the client put it on the first call.
WHAT WE DIDWe agreed what “healthy” means for each service, instrumented logs, metrics and traces to match, and rebuilt alerting around it.
15 Security & reliability · SRE & incident responseEuropean payments company On-call is shared fairly, incidents follow a known routine, and each one leaves a written lesson.
CHALLENGEThe same two people get paged every time. That was how the client put it on the first call.
WHAT WE DIDWe set service level objectives with the product owners, restructured the on-call rota, and introduced a simple incident process with written reviews.
16 Security & reliability · Disaster recovery & backupLondon professional services SME A recovery plan with timings measured in a real test, repeated on a schedule.
CHALLENGEWe’ve never tested a restore. That was how the client put it on the first call.
WHAT WE DIDWe agreed recovery targets with the business, rebuilt backups to meet them, and ran a full restore test with the client’s team.
17 Efficiency & innovation · Cost optimisationUK e-commerce scale-up A lower bill, a monthly cost report per team, and budgets that engineering leads own.
CHALLENGEThe cloud bill was rising faster than traffic, and nobody could say which product was responsible.
WHAT WE DIDWe tagged and attributed spend by team, removed idle resources, rightsized compute and databases, and set up commitments for the steady load.
18 Efficiency & innovation · AIUK B2B software company One AI feature in production, with data-handling rules the client’s security lead signed off.
CHALLENGEWe want AI in the product but don’t know where to start. That was how the client put it on the first call.
WHAT WE DIDWe ran a short discovery to pick one use case, built search over the client’s own documents with access controls, and shipped it behind a feature flag.