Agentic coding is coming to FileMaker, and plenty of us aren't waiting. We're already using AI to generate code and paste it into client files.
However, using AI doesn't replace the need for best practices, and if anything, it makes them matter more. FileMaker makes those practices harder to follow than most development environments, because there's no diff, no pull request, and no revert. And our clients are trusting us to work inside systems their business depends on, which raises a fair question: what should they be able to expect from us, and what should we be able to show them?
Demoing in Claude Code, I'll work through the practices and guardrails that answer it, grounded in responsible AI principles and in frameworks like the NIST AI Risk Management Framework and ISO/IEC 42001. They come down to privacy, security, reliability, and accountability, translated into things we can put in place and, when a client asks, point to.
After two decades of FileMaker development, Kate Waldhauser founded Violet Beacon when it became clear that helping organizations use AI responsibly was work worth doing. She's an ISO/IEC 42001 Lead Implementer, which means she works at the intersection of FileMaker development and AI governance and has opinions about where the two collide. She writes Human Intelligence, a blog and newsletter offering non-hype thinking about AI, and works from San Marcos, Texas.
Presenters:
Kate Waldhauser
Chief Strategist
Violet Beacon
Schedule:
- Meet and Greet (Time 11:00 am)
- Presentation (11:15 am)
- Q & A and Discussion (12:00 pm)
- Lunch (12:30 pm)
