Freelance AWS consultant

Most of my work over the last decade has been on AWS, for an online bank, an international leasing company and a manufacturing marketplace. These are the jobs I usually get called in for.

Scope

What I take on

Account and network layout

Multi-account structure, VPC and subnet design, segmentation, and the dull guardrails that stop an estate sprawling, so it still makes sense in two years.

Workloads

ECS and EC2, RDS, S3, DynamoDB, API Gateway, CloudFront. Sizing, placement, autoscaling, and the ways each of them actually fails.

Migration off legacy

Moving estates onto AWS from on-prem, from a single monolithic environment or from a hosted provider. In stages, with production running the whole time.

Cost and availability

Finding the money going on idle capacity, and the single point of failure that is waiting for a bad Tuesday.

Method

How it runs

  1. 01

    Read the estate first

    Before anything moves, I map what is really running, what depends on what, and which parts nobody dares touch. Migration plans built from the architecture diagram instead of from what is actually running are how weekends get ruined.

  2. 02

    Codify, then cut over

    The target gets built in Terraform before any traffic goes near it. That makes the cutover reversible, and your team ends up with infrastructure it can rebuild instead of a hand-made environment nobody understands.

  3. 03

    One environment at a time

    Staged migrations with a real rollback path. I have never needed a big-bang weekend cutover and would push back hard on one.

Evidence

Where I have done this before

  • 3D Hubs Redesigned a monolithic infrastructure into isolated, reproducible environments on AWS with Terraform, so scaling and testing stopped getting in each other's way.
  • LeasePlan Ran SAP OpenText hosting on AWS where high availability was the hard requirement, and automated the instance configuration that had been done by hand.
  • Knab AWS work inside a regulated Dutch bank: ECS, RDS, CloudFront, API Gateway and DynamoDB, with the evidence requirements that come with financial services.

Fit

Probably worth a conversation if

  • You are on AWS already, but it was built by hand and nobody can reproduce it.
  • You are planning a migration and want someone who has done staged cutovers before.
  • Your bill or your availability says the architecture needs a rethink.
  • You need AWS depth for a fixed period without hiring a permanent engineer.

Contact

Tell me what is actually broken

The contact details live on the front page. If I am not the right person for this, I will say so.