West Valley Flying Club uses Schedule Master for its scheduling. It’s solid software, but it has gaps that can leave instructors and members out of sync. A member has trouble booking regularly, life gets in the way, or there’s some other complication that keeps them from flying. Time passes. That’s not good for CFIs, members, or the club.
I’m building a pipeline that fixes this. It pulls member activity data, scores each person by engagement urgency (how long since their last flight, whether they’ve gone quiet), and matches them to their flight instructor. Each morning, students get an email about when and who to book with to stay on track, where they are in their training, and what steps to look for next. Instructors get the same visibility from their side, so no one falls through the cracks.
Built with Python and n8n. The scoring logic doesn’t need a model, just clear rules about what “drifting” looks like in a flying club context.
I kept running into the same problem with prompts: I’d write one, it seemed to work, I’d move on. But I never really knew if it was actually better than the version I didn’t try.
So I built an A/B testing system, using aviation weather as the testing ground. It pulls live METAR and forecast data, runs two prompt versions against the same real conditions, and scores each response with an LLM judge on accuracy, clarity, actionability, and tone. Results build up across multiple trials so you’re comparing patterns, not a single output.
I wanted something I’d actually trust for picking prompts. A repeatable process I could point to and say, this one’s better, and here’s why.
- The Blindspot — Issue No. 1 May 2026
- Aviation Substack ongoing
I’m a solutions consultant focused on AI applications based in Mountain View, CA. At West Valley Flying Club at Palo Alto Airport, I build automation tools, support the safety program, and help write for a safety newsletter called The Blindspot. My background is in linguistics, education, and prompt engineering. I spent time on Meta’s Gen AI team doing red teaming, model evaluation, and technical writing. I think a lot about how language and systems interact. How you structure a prompt. How you design a workflow. How you explain something complex so it actually lands.
I’m also a student pilot working toward my private certificate. I believe that flying teaches you what it means to build tools people depend on, and what good checklists, clear communication, and solid systems actually look like when the stakes are real.
I’m looking for roles in AI deployment, prompt engineering, or solutions architecture, especially with small teams doing work that matters.
