I work in platform operations, with a focus on Linux, cloud infrastructure, distributed workloads, monitoring, automation, and production troubleshooting.

I build tools for inspecting systems, diagnosing failures, reducing operational overhead, and making services easier to run and maintain.

My background is in scientific computing and data-intensive platforms, where I have worked with AWS, Google Cloud, DNAnexus, Apache Beam, Spark, Slurm, Docker, systemd, and HPC environments.

Currently open to platform operations, cloud operations, production support, systems engineering, technical support engineering, and scientific infrastructure roles.

• Linux service and process monitoring
• Production incident investigation and root-cause analysis
• Cloud workload troubleshooting across AWS and GCP
• Automation with Python, Bash, Go, and Rust
• Distributed runtime and dependency diagnostics
• Systemd services, structured logging, metrics, and health checks
• Resource usage, reliability, and cloud cost optimisation
• Operational runbooks and reproducible failure investigation

Systems & Reliability: Linux/Unix, systemd, cgroups, process trees, /proc, CPU, memory, disk I/O, file descriptors, TCP/UDP, Unix sockets, incident investigation

Cloud & Infrastructure: AWS, Google Cloud Platform, IAM and permissions, cloud compute and storage, Terraform/OpenTofu, Docker/Podman

Monitoring & Operations: OpenMetrics, structured JSON logs, HTTP health checks, runtime diagnostics, process monitoring, resource thresholds, operational runbooks

Distributed Systems: Apache Beam, Spark, Slurm, HPC, DNAnexus, Nextflow, batch execution, job scheduling, worker/runtime troubleshooting

Languages & Tooling: Python, Bash, Go, Rust, SQL, Git, GitHub Actions, Makefile, REST APIs, CLI development

A Linux service and process monitor written in Python.

RunWatch can inspect systemd services, process trees, cgroups, CPU, memory, disk I/O, file descriptors, network connections, Unix sockets, and HTTP endpoints.

A dependency-risk and upgrade-impact inspection tool for PyPI and Conda/Bioconda environments.

It traces transitive dependencies, explains package relationships, estimates upgrade blast radius, identifies environment conflicts, renders dependency trees, and caches repeatable analysis results.

A lightweight Go command runner that watches files and reruns commands when changes occur.

Supports recursive watching, glob and standard-input selection, configurable output handling, exit-code propagation, testing, and installation tooling.

A Go CLI for semantic commits and changelog automation.

My terminal-first Linux development environment, including shell, Neovim, and Tmux configuration.

Reliable systems are not defined by whether they work once.

They should be:

• observable
• reproducible
• easy to inspect
• safe to operate
• predictable under failure
• explicit about limitations
• recoverable when deployment goes wrong
• understandable by the next engineer

I like the engineering work that happens between "the code runs" and "the system can be trusted."

LinkedIn · GitHub · Email · Buy Me a Coffee