Systems and cloud infrastructure engineer in London. MEng from UCL (dissertation on automatic parallelisation across heterogeneous computing systems). I care about what happens beneath the abstraction — how bytes move, how kernels schedule, and why that last microsecond matters.

I build tools in Go, tinker with Linux internals, and run a self-built production Kubernetes cluster on Hetzner Cloud with kubeadm, ArgoCD, Prometheus/Grafana and Terraform-automated provisioning.

I also experiment with AI agents especially around the reliability space to improve and automate debugging, scaling and provisioning of infra.

Linux internals — virtual memory management, filesystem layer (VFS, page cache, inode/dentry), cgroups and namespaces

Go systems programming — built a FUSE filesystem with content-addressed storage, terminal UIs, cross-platform system monitors

Kubernetes & cloud infrastructure — cluster bootstrapping with Terraform, GitOps with ArgoCD, CNI plugins, eBPF/Cilium

Observability — designed distributed tracing pipelines with OpenTelemetry, built Prometheus/Grafana/Alertmanager stacks

Low-latency systems — explored memory pre-allocation for HFT workloads, NUMA-aware design, FUSE vs native filesystem performance

Python & Bash — async patterns, GIL-aware concurrency, infrastructure automation scripting