What I build
I like building tools that sit close to real workflows: capture helpers, formatting utilities, automation concepts, learning systems, and developer-focused interfaces that turn rough input into something structured and useful.
About
I build software that makes everyday work easier.
I build software and automation around real tasks. My work is language-agnostic: I choose the tools that fit the problem and use AI to help build, inspect, and revise the result. This site is a place to collect public projects, small utilities, and notes from experiments that are safe to share. A lot of my current work centers on Brainwav: my personal system for capturing knowledge, keeping project context, and automating recurring work. Codex handles much of the hands-on building and revision work, with other tools supporting the workflow.
Current interests
- AI-assisted development
- Practical calculators and browser tools
- Choosing technology around the task
- Automation
- Repository-aware AI workflows
- Agent runtimes and local tooling
- Learning systems
How I Work
I start with the task, keep the scope clear, and check the result.
Start with the task
Start with the problem, then choose the simplest system that can solve it well.
Focused tools
Build focused utilities that remove repeated friction from real workflows.
Maintainable systems
Favor maintainable structure, explicit tradeoffs, and code that is easy to revisit.
Verify and improve
Get a useful version working, verify it, and improve continuously based on what matters.
Examples of my work
The Detailer Calculator continues a Windows tool as a mobile-friendly web app. Jones Building Takeoff handles building-level material quantities. My Python journal and Vector notification service solve different personal workflows.
These projects span several languages. The common thread is the problem being solved and the checks that make the result useful.