Engineering beyond the application
At Flight Level Engineering, where I have worked as a Software Engineer since approximately April 2023, my responsibilities span software architecture, system integration, and the infrastructure that helps engineering teams deliver their work.
I designed and implemented a Jenkins-based CI/CD environment from the ground up, including reusable pipelines and shared libraries supporting Windows and Linux. This was infrastructure creation—not simply using a pipeline someone else maintained.
I also served as the primary software lead for delivery of a complex helicopter simulation system. That work involved end-to-end software ownership, coordination across engineering disciplines, integration, and troubleshooting at the software/hardware boundary.
The common thread: ownership
C++ and Python are central to how I work. But the language is only one part of the problem. A system also needs clear communication interfaces, reliable connectivity, reproducible builds, practical deployment, and a way to diagnose unexpected behavior.
My work with real-time control-loading software, configurable UDP communications, and engineering networks has made that cross-layer perspective essential. I want to understand where a problem actually comes from, rather than assuming it belongs to one component.
Measure. Simplify. Automate.
My approach is to make constraints explicit, keep interfaces understandable, and use evidence to guide improvements. For performance-sensitive work, that means profiling, identifying bottlenecks, benchmarking changes, and considering resource trade-offs rather than optimizing by intuition.
For recurring engineering work, it means building automation that other developers can understand and maintain. The aim is not just to make one task faster, but to make the next delivery more dependable.
Personal engineering R&D
Outside professional work, I experiment with self-hosted language models, local inference, GPU resource constraints, and autonomous developer tooling. This includes llama.cpp, Ollama, long-context inference experiments, MCP integrations, and agent orchestration.
These are personal research projects and prototypes. They are not presented as commercial AI deployments or evidence of production-scale AI consulting.
What I’m interested in discussing
Difficult C++ applications, engineering automation, CI/CD infrastructure, networking software, real-time integration, and technical roles where understanding the entire system matters.
Explore the engineering work behind these capabilities, or get in touch about your problem.