Hey, I'm Wak Bijak.
I have spent twelve years building infrastructure: networks, servers, storage, and the platforms that live on top of them. A good chunk of that was rescue work: walking into a stalled multi-year program, becoming the architect across infrastructure, software, and delivery because somebody had to, and walking out with running costs cut roughly in half and a permanent habit of asking why everything costs so much.
These days the day job is designing AI datacenters and enterprise infrastructure: GPU compute, network fabrics, storage that has to survive audits. The night job is a small Kubernetes cluster at home where I test ideas before I recommend them to people who pay for certainty.
What I spend my time on:
- Cloud & Infrastructure: designing systems that stay up, and knowing exactly why when they don't.
- DevOps & Platform Engineering: Kubernetes, gitops, CI/CD, and the boring automation that makes everything else possible.
- AI Engineering: agents, memory systems, AI datacenter design, and getting LLMs to do useful work instead of just demos.
- System Integration: making vendor A talk to vendor B without anyone crying.
I also teach when I can: training modules and hands-on labs, because the fastest way to find the holes in your knowledge is to explain it to someone else.
Most of what I write here comes from things I actually ran, broke, and fixed. Some of it in production, a lot of it in that cluster at home. I write things down mostly so future me stops repeating past me's mistakes. If it helps you too, even better.
Based in Malaysia. Posts are casual, opinions are mine, uptime is mostly the cluster's.