Outcome 03
Modernise your platform
Platforms accumulate decisions. I come in to migrate, stabilise, and right-size — so the platform supports the business instead of fighting it.
The symptoms
- The platform is hard to change and incidents are frequent.
- Cloud spend is rising faster than usage justifies.
- There’s little observability — you find out from customers.
- A migration has stalled, or never started.
What I do
- 01Run an architecture review that names the real risks and a realistic path to fix them.
- 02Lead cloud migrations and platform rebuilds — Kubernetes, GitOps, infrastructure as code.
- 03Put observability and reliability practices in place so problems surface early.
- 04Bring cloud cost under control without slowing the team down.
Proof
AI/ML platform engineer — e-commerce group
Built a cloud-native platform on AWS EKS with Terraform and FluxCD GitOps, including deduplication, dead-letter queues, and S3-backed disaster recovery — on a few hundred dollars of infrastructure a month.
GitOps, IaC, DR by design
Fractional CTO — agritech SaaS
Delivered a comprehensive architecture review with board-level technology strategy and cost reporting, and began realising savings through logging reduction and compute right-sizing.
25–50% Azure savings identified
Co-founder & CTO — ML personalisation
Ran a real-time personalisation platform for major broadcasters as a two-person company — canary releases, Prometheus monitoring, and a formal postmortem practice, under contractual SLAs.
10k+ req/s at 19 ms p99
Lead data engineer — broadcaster
Built a Lambda-style architecture combining batch and real-time pipelines on Spark and Kafka at high throughput.
50+ TB per day