About Me

I'm Dan Foulkes — a Site Reliability Engineer with a passion for building things that run reliably at scale, and then rebuilding them at home to understand why.

This blog is where I write up the projects I've been running in my homelab: Kubernetes clusters on Raspberry Pis, observability stacks, AI experiments on AWS Bedrock, and the occasional tangent into Linux, networking, or whatever I've broken recently.

What I work on

My day job is SRE work — keeping distributed systems healthy, thinking about failure modes, and building the tooling that lets engineers sleep at night. Outside of work, I run a homelab that acts as a testbed for the same ideas: a Raspberry Pi K3s cluster, a Proxmox hypervisor, and an observability stack built around Prometheus, Loki, and Jaeger.

More recently I've been building AI tooling — including Avengers Assemble, a council of specialist AI models running on AWS Bedrock inside Kubernetes that deliberates on hard questions by fanning them out across multiple models in parallel.

Tech I use

  • Infrastructure: Kubernetes (K3s), Proxmox, Terraform, Helm
  • Networking: Traefik, MetalLB, Cloudflare, cert-manager
  • Observability: Prometheus, Loki, Jaeger, OpenTelemetry
  • Cloud: AWS (Bedrock, S3, IAM)
  • Languages: Rust, TypeScript, Go, Python
  • CI/CD: GitHub Actions, pre-commit

Writing

I write to learn. Working through something well enough to explain it clearly is the best way I know to find out what I actually understand versus what I only think I understand. If a post has helped you, I'd love to know.

Posts go out irregularly — when something is interesting enough to write up, not on a schedule. You can follow along via the RSS feed.