Technical services
Cloud & Infrastructure
This is the ground the product stands on. We design the environment, automate the way code reaches it, and keep the running system patched, backed up, and watched. Everything here is configured in your own accounts, so the bills come to you directly and nothing is locked behind us. We will also tell you plainly where our control ends — the hosting provider's network is not ours, and neither is a password an employee gives away.
Cloud Architecture
The virtual data centre your application lives in, sized for the traffic you actually have and able to absorb the spikes you can name. Most cloud bills are large because nobody chose the shape of the environment — servers were added, never removed. We design it before provisioning it, on AWS, Azure, or Google Cloud, picked for what you run rather than for what we prefer.
What you get
- A configured environment on AWS, Azure, or Google Cloud, inside your own account
- Compute sized to the load you describe: virtual machines or Kubernetes, with serverless functions where they genuinely fit
- Managed databases with automated backups and a restore path that has been written down
- Private networking, security groups, and access rules — nothing exposed that does not need to be
- An architecture document: what runs where, what drives the monthly bill, and which part will bend first under load
What we need from you
- Expected traffic, data volume, and any spike you can predict — a launch, a campaign, a season
- Your own cloud account, or a decision to open one, because the bill is yours
- Any data residency or compliance requirement, before the design, not after it
- Access to the current environment and its configuration if you are migrating
Where it stops
- Cloud costs are billed by the provider to your own account. We size and configure; we do not resell capacity or add a margin to it
- We can remove waste, but no one can promise you a monthly figure — your usage sets the bill
- Migrating a live production system is scoped separately from designing the environment
- A configured environment that nobody watches is not managed infrastructure. Running it is a Care Plan or a DevOps engagement
- Third-party licences, managed service fees, and data transfer charges are your own cost
DevOps & CI/CD
An automated assembly line for your code: every change is built, tested, and deployed the same way, by a script instead of by whoever remembers the steps. The point is not speed for its own sake. Teams that can deploy in minutes ship smaller changes, and small changes are the ones you can undo. So the rollback path gets built and tested before the first release, not improvised during the first outage.
What you get
- Automated pipelines on GitHub Actions or GitLab CI that build, test, and deploy on every change
- Separate development, staging, and production environments, so nothing is tested on the live system
- Infrastructure as Code with Terraform or CloudFormation, so an environment is rebuilt from a file rather than from memory
- Container configuration (Docker), so the application behaves the same on every machine it lands on
- A written and tested rollback path, plus deploys that do not take the site offline where the architecture allows it
- Documentation of the pipeline, the secrets it needs, and who is allowed to release
What we need from you
- Repository access, and a branching model — or a decision to adopt one
- Credentials for the cloud or hosting account the pipeline will deploy into
- A decision on who approves a production release
- The current build and deploy steps, including the undocumented ones somebody does by hand
Where it stops
- MVP Build includes a basic pipeline for the product we deliver. A full DevOps engagement is quoted separately
- Rewriting an application so it can be deployed automatically — splitting a monolith, removing hard-coded configuration — is its own project, not a pipeline task
- Day-to-day release operations stay with whoever owns the code, unless a monthly plan covers it
- CI runner minutes, container registry storage, and tool licences are billed to your own account
- Pipelines age. When the tooling or the platform changes and no plan covers it, nobody is keeping them working
Hosting & Security
We manage the domain, the servers, and the gates: SSL, DNS, firewall rules, DDoS filtering, backups, and updates on a schedule instead of when someone remembers. Most sites that get broken into were not targeted — they were running a plugin with a published vulnerability for eight months. Patching on a rhythm is the whole trick, and it is dull work that has to happen anyway.
What you get
- Managed hosting with SSL and DNS configured, and a live https address
- DDoS protection and firewall rules that stop automated traffic before it reaches the application
- Scheduled backups with retention by tier — weekly and 30 days on Launch, daily and 90 days on Grow, daily plus on-demand and 365 days on Scale — and restores on request
- Core, plugin, and dependency updates: monthly on Launch, weekly on Grow, weekly plus emergency security patches on Scale
- A staging environment from Grow up, so a change is tested before it touches the live site
- Uptime monitoring: 5-minute checks with email alerts on Launch, 1-minute checks on Grow, and 1-minute checks with 24/7 automated alerting and escalation on Scale
- Malware scanning on every tier; clean-up and recovery from Grow, and a hardening review twice a year on Scale
- A monthly health report
What we need from you
- Access to the domain registrar and the DNS records
- Admin access to the current hosting and to the site or application itself
- An agreed list of everyone else who holds admin access
- One point of contact who can approve an urgent change without a meeting
Where it stops
- Uptime depends on the hosting provider's own availability. We monitor, report, and escalate; we do not control their network. Launch and Grow are best effort. Scale carries a 99.9% monthly target that we report against — a target, not a guarantee
- 24/7 alerting on Scale is automated monitoring with escalation. Human response times are stated in business days: same business day on Scale
- We secure the infrastructure. We cannot secure a password an employee hands over — social engineering is outside the reach of any firewall
- Recovery from changes made by someone else with admin access is not covered by the plan's included hours
- We cannot take responsibility for infrastructure we did not build or cannot access
- Hosting resources beyond your plan's allowance, domain fees, and third-party licences are billed to your own account
- New pages, new features, and redesigns are not maintenance. They are quoted separately
Where these are sold
Hosting and security are the Care Plan, in three tiers — attached at the end of every build and sold on its own for a site that already exists. Cloud architecture and full DevOps engagements are Custom Solutions, quoted after a paid scoping step: an AI & Systems Audit for an operations problem, a Discovery Sprint for a product. MVP Build already includes a basic pipeline for what we deliver. Every build ends with a 30-day Hyper-Care warranty on delivered scope; after that a Care Plan takes over.