Experience

Employed full time to work on open source, mainly Flatcar Container Linux, a CNCF immutable operating system for container and Kubernetes workloads at scale. I also work on Azure internals such as provisioning. Being paid to contribute upstream full time has meant learning how to work in the open with a global community, balancing long term project direction with the day to day realities of maintaining a production grade Linux distribution used across the industry.

Built risk and exposure tooling for a live trading platform, including AI-driven detection of high-risk trading activity and automated regulatory reporting pipelines on Spark and GCP. Also mentored other developers on architecture and clean code. Working so close to live financial markets meant getting comfortable with strict regulatory requirements and the kind of reliability expectations that come with systems handling real money, which sharpened how I think about correctness and risk in everything I build.

Joined as one of the founding team members and led development from the ground up, putting my security, compliance, and cloud engineering skills to direct use. Being there for the early days of a startup taught me a lot about the business side: talking to VCs and other CEOs gave me a real look at what it takes to get a company off the ground. I left when the product direction narrowed to banking compliance only, since I wanted to keep growing on the technical side instead.

Built automated video anomaly detection for cinema systems and set up CI/CD pipelines that the wider team relied on, working mostly in Python and C on Linux based cinema hardware. It was my first real exposure to shipping software that had to work reliably in physical cinemas around the world, which meant paying a lot of attention to edge cases and making sure automated checks could catch problems long before they reached a real projector.

Improved Dolby's subtitle rendering engine and led development of synchronisation tooling used across multiple cinema projectors, writing C and Python on Linux systems throughout. Working on tooling that other engineers and cinemas depended on daily taught me how much low level timing and rendering details matter when the audience is sitting in a dark room expecting everything to just work.