gremlin.group / services
What we do.
We have been building, fixing, and migrating infrastructure for over 20 years. Here is where we spend most of our time.
AI & Machine Learning
Most useful AI is not a chatbot on your website. It is document processing, knowledge search, automated reports. The kind of thing that turns a full day of manual work into half an hour of checking edge cases.
How we approach it
We start with the business problem. If a spreadsheet formula solves it, we will tell you. If you need a proper ML pipeline, we build it with testing, monitoring, and a plan for what happens when the model gets something wrong.
- ML model development and training pipelines
- Natural language processing and document extraction
- Computer vision for quality control and automation
- AI-powered search and recommendation systems
- LLM integration and prompt engineering
- Model monitoring, evaluation, and retraining workflows
AWS & Cloud Architecture
AWS is where we spend most of our time. Single-account setups for startups, multi-region architectures handling NHS data residency, and everything in between. If your infrastructure is costing too much or not scaling the way you need, we can usually spot the problem quickly.
How we approach it
We look at what you have, work out what you actually need, and build that. Sometimes that means a few targeted changes instead of an expensive redesign.
- Infrastructure as Code with Terraform, CloudFormation, and Pulumi
- Cost optimisation audits that typically save clients 30-50%
- High availability and disaster recovery planning
- Security hardening and compliance (SOC 2, ISO 27001)
- Serverless architecture with Lambda, API Gateway, and Step Functions
- Monitoring and alerting with CloudWatch and custom dashboards
Cloud Migration
We have moved everything from servers under desks to full data centres. The first thing we do is find out what you actually have, because there is always a system running somewhere that nobody remembers setting up. Then we plan a phased move so nothing breaks on the way.
How we approach it
We move the least critical things first to prove the process, then work up to the important stuff. Every step has a rollback plan. We run both environments in parallel until we are confident the new one is solid.
- Full discovery and dependency mapping before anything moves
- Lift-and-shift, re-platform, or re-architect depending on what makes sense
- Database migration with minimal downtime
- Application modernisation where it saves you money long-term
- DNS and networking cutover with rollback at every stage
- Post-migration optimisation and cost analysis
Kubernetes & Containers
Container orchestration is powerful, but it is not always the right call. When it is, we get clusters running properly and hand them over so your team can manage them day to day.
How we approach it
We start with whether Kubernetes is the right fit. If it is, we set up a cluster your team can understand and run. If ECS or Cloud Run would serve you better, we will say so.
- Cluster design and deployment on EKS, GKE, or AKS
- CI/CD pipeline integration with automated deployments
- Helm chart development and management
- Monitoring with Prometheus and Grafana
- Security policies, RBAC, and network policies
- Cost optimisation through right-sizing and spot instances
Software Development
APIs, web applications, internal tools. Clean code that someone else can pick up and maintain after we leave. Anyone who has inherited a messy codebase knows why that matters.
How we approach it
We build iteratively and ship frequently. You see working software early, not a slide deck. Everything is tested, documented, and version controlled from day one.
- Full-stack web applications with modern frameworks
- REST and GraphQL API design and development
- Microservices architecture and implementation
- Database design and optimisation
- Automated testing and CI/CD pipelines
- Technical debt reduction and legacy modernisation
Charity IT Support
We give charities our time at reduced rates, and some get it for free. We have helped organisations move off servers under desks, set up proper backups, and get their teams on Microsoft 365 for next to nothing using charity licensing.
How we approach it
We focus on the changes that make the biggest difference first: getting your data backed up, your accounts secured, and your team able to work without fighting the technology.
- Microsoft 365 setup and management (charity licensing)
- Cloud migration and data backup strategies
- Cybersecurity basics: MFA, device management, training
- Website hosting and maintenance
- IT strategy and roadmap planning
- Ongoing support with priority response times
Not sure which of these you need?
Tell us what is going on and we will give you an honest answer, including when the answer is not us.