Software Engineer · Computer Science PhD Student · Systems Tinkerer
I like building things, taking systems apart to understand how they work, and occasionally turning small problems into unnecessarily elaborate side projects.
My research is mostly around distributed systems, storage, key management, security, and performance. Right now, I work on distributed KMS architectures and look at how they behave under load, scaling pressure, and failure.
Outside of research, things get a little less focused.
I spend a lot of time experimenting with Linux, NixOS, self-hosting, local AI, developer tooling, infrastructure, automation, and whatever technical rabbit hole seems interesting that week.
A surprisingly common pattern is:
Find a tool → try it → find something annoying → try four alternatives → wonder if I should just build it myself.
- 🔐 Distributed KMS architectures, performance, scaling, and failure behavior
- 🛠️ Ground Control, a control plane for heterogeneous infrastructure
- 🐧 Linux, NixOS, reproducible environments, and infrastructure experiments
- 🧠 Local LLMs, AI agents, MCP/LSP tooling, and benchmarking
- 🏠 Self-hosting, homelab projects, automation, and small tools that may or may not have needed to exist
Languages & frameworks
Systems & infrastructure
Data & databases
backend services · research prototypes · developer tools · automation · infra experiments · benchmarks · dashboards · things I got tired of looking for
Some projects are useful.
Some are experiments.
Some exist because I spent two hours looking for the right tool and decided hour three should apparently be spent writing one instead.
- distributed systems and storage
- backend architecture and APIs
- Linux and reproducible environments
- security tooling and vulnerability management
- performance testing and benchmarking
- self-hosting and infrastructure
- local AI and coding agents
- open source and developer tooling
Most of my actual coding is in Go and Python, with varying amounts of C#/.NET, TypeScript, SQL, Docker, Linux, and AWS mixed in.
If something can be benchmarked, automated, self-hosted, customized, or taken apart to understand better, there's a decent chance I'll end up doing exactly that.
Building things, breaking things, and occasionally pretending that was the plan.