About
I'm a self-taught infrastructure engineer who builds, deploys, and maintains production-grade AI systems. Not on AWS or Azure — on bare metal. The skills are the same; the difference is I built it all from scratch.
I design and maintain the infrastructure that runs large language models — the plumbing between AI models and the people using them. Multi-provider model routing, GPU inference servers, voice pipelines, monitoring, cost optimisation. All containerized, all automated, all running 24/7.
I don't inherit working stacks. I build from scratch, break things, fix them, and iterate. No ticket queue, no manager — just making things work. Self-directed learning, continuous improvement, and a deep understanding of how each layer of the stack connects.
Container orchestration (Docker), Linux systems administration, Python scripting, GPU inference (llama.cpp, vLLM), model routing (LiteLLM, OmniRoute), observability, automation, networking, security. I can read code, debug systems, and deploy production services.
Kubernetes (pods, deployments, services), Terraform (infrastructure as code), cloud platforms (AWS/GCP). I've been told these are gaps — and I'm actively closing them. Two to four weeks to productive with K8s; the concepts transfer directly from Docker Compose.
I'm looking for Platform Engineer, AI Infrastructure Engineer, or DevOps roles. Remote-first, freelance or full-time.
Get in Touch