Services
Three kinds of work. Senior people doing it.
Platform, AI adoption and software engineering, sold as our consultants' time rather than as packages. Scope, team and dates are agreed in writing before anything starts.
Platform
Platform, AWS and SRE engineering
Most of our consultants' careers have been inside AWS estates: building them, running them, and finding out at three in the morning what was wrong with them. We work in your accounts, with your team, and leave infrastructure as code behind.
Typical work
- Well-Architected, cost and security reviews
- Multi-account landing zones and governance
- Migrations, with AWS MAP funding where the workload qualifies
- Terraform and CDK
- Observability, on-call and incident response
AI
AI adoption
Picking the use case that pays, proving it on your data, and running it in production with a bill you can read. Your systems stay where they are. The model runs on infrastructure we set up and manage.
Typical work
- Use-case selection and data readiness
- Proof of concept on your own data
- Amazon Bedrock, guardrails and evaluation
- Running-cost modelling
- Production hardening and handover
Working examples
What AI actually does in a small business
Not robots, not magic. Just software that takes the repetitive part of a job and gives the hours back. These are the kinds of things a short proof of concept settles.
Answer the same questions, once
Now Your inbox fills with the same twenty questions every week.
With AI An assistant drafts each reply from your past answers. Your team checks and sends.
Paperwork that types itself
Now Someone retypes invoices, delivery notes and contracts into the system.
With AI Amounts, dates and terms land in your spreadsheet or accounts package on their own.
Ask your own documents
Now The answer is in a manual, a policy or an old proposal. Somewhere.
With AI Ask in plain English, get the answer back with the source it came from.
Quotes built on past jobs
Now Estimates start from a blank page and whoever wrote the last one.
With AI A draft quote based on what similar jobs actually cost you, ready to adjust.
Your systems stay where they are, whether that's on-prem, Microsoft, Google, or a folder of spreadsheets. We've built on AWS for a decade. You never have to log into it.
Software
Software engineering
Backend and platform code, shipped into your repository with tests and a pipeline. We are as happy paying down the code nobody wants to touch as building the new thing.
Typical work
- Backend services in TypeScript and Python
- APIs and event-driven systems
- CI/CD pipelines
- Modernising legacy code
- Code and architecture reviews
Engagements
How engagements work
Every engagement is our consultants' time, scoped in writing before it starts. Three shapes cover most of them.
-
Embedded
01One consultant in your team
A senior engineer working inside your team, in your tools and your stand-ups, for as long as the work needs.
-
Two to four consultants
02A small team for a build
A lead and a small team, a defined deliverable in your repositories and accounts, and a written update every week.
-
A set number of days a month
03Ongoing days
Consultant days for architecture, reviews and escalation when something breaks. Month to month, cancel with notice.
The proposal states who, how many days, the deliverable and the end date. Nothing else is billed.
Case studies
Real engagements, with numbers
Platform, AI and software work, with the numbers that came out of it.
Acorn (Compliance Platform)
HealthTech / SaaS
One consultant in your team
Zero-downtime launch, NHS DSPT due diligence passed
Acorn needed to launch 'Squirrel 2,' a critical compliance platform for the NHS and healthcare providers. As an incredibly cost-conscious company, they required a secure, scalable infrastructure that could move off expensive EC2 instances while meeting strict NHS compliance standards.
- Successful Launch of Squirrel 2: The platform went live with zero downtime, immediately serving NHS and private healthcare clients
- 100% Compliance Confidence: The architecture successfully passed strict due diligence, enabling Acorn to sell into highly regulated NHS trusts without friction
- Scalable Foundation: The new infrastructure automatically adjusts to traffic spikes, ensuring performance never degrades for compliance officers using the tool
Alluvial (Institutional Liquid Staking)
FinTech / Web3
A small team for a build
Passed multiple external security audits
Institutional investors needed a compliant way to stake assets. They required the yield of staking but needed the flexibility to toggle between 'Private' (compliant) and 'Liquid' (public) states based on shifting regulatory landscapes.
- Regulatory Flexibility: Provided a technical solution to a legal problem, allowing the product to launch despite regulatory ambiguity
- Audit-Ready Code: The strict CI/CD and typed Go infrastructure ensured the codebase passed multiple external security audits with flying colours
- High Availability: The Kubernetes-based off-chain layer ensured that data feeds and APIs remained responsive even during periods of high network congestion
EDSports (EdTech Platform)
Education / Gaming
A small team for a build
Setup time cut from 20 minutes to 30 seconds
EDSports wanted to bring Minecraft Education esports to schools, but most schools rely on low-spec Chromebooks that cannot run the game locally. They needed a zero-install, GDPR-compliant solution.
- 100% Hardware Agnostic: High-fidelity Minecraft now runs smoothly on $150 Chromebooks and ten-year-old laptops
- Instant Access: No installation or patching required by school IT admins. Students simply log in and play
- Teacher time back: The custom launcher reduced setup time from 20 minutes to 30 seconds, maximising learning time
Facing something similar?
Book a callFAQ
Questions we hear a lot
What happens on the first call?
Thirty minutes.
You describe the problem, we ask questions, and we work out whether there is a fit. If there isn't, we'll say so on the call and point you somewhere better if we can. No deck and no follow-up sequence.
How does an engagement start?
With a written proposal.
It states the scope, who is on the team, how many days the work needs, and the end date. Nothing is billed until you've agreed it.
Most engagements take one of three shapes: one consultant embedded in your team, a small team for a defined build, or a set number of consultant days a month for ongoing work.
Who does the work?
Senior consultants.
The proposal names who is on the engagement, and you can meet them before it starts. We don't swap people in without telling you. When a piece of work needs another specialism, we bring in a consultant who has it, and the same rule applies.
Do you only work on AWS?
No.
AWS is where most of our platform work has been, and North Point Digital is an official AWS Partner. AI adoption and software engineering don't depend on where your systems live.
For AI work, your data stays where it is: on-prem, Microsoft, Google, or spreadsheets and email. The model runs on infrastructure we set up and manage, and you never need to log into it.
Do you work remotely or on site?
Remotely by default, from the UK.
Our consultants work in your chat tools, your repositories and your accounts. Days on site for workshops or kick-offs are fine where they're useful, and they go in the proposal like everything else.
What happens at the end?
Handover.
Documentation, infrastructure as code and a knowledge-transfer session, so your team can run what we've built without us. There's nothing to cancel afterwards. Clients who come back do so because they want to, not because they're locked in.
Bring the problem. We'll tell you how we'd staff it.
Thirty minutes. If it is not a fit, you will hear that on the call.
Official AWS Partner
Scope, team and dates in writing before work starts Nothing to cancel afterwards