2nd Event - AI-First Engineering with Claude Code

Join us in Irvine for a second hands-on, developer-focused evening on how AI is changing the way we build software—today, not someday. Team Artemis, the AI-first innovation group at PlanetDDS, hosts this meetup.
We’ll explore how Claude Code is speeding up development and what Compound Functional Requirements looks like in practice. You’ll see how to build a knowledge base using source code.

What to expect:

• Registration: 4:30 - 5:00 PM
• How we use Claude Code (60 min)
• Networking & Happy hour (60 min)
  Meet other engineers, founders, and AI builders. Swap ideas, compare workflows, and talk honestly about what’s working (and what isn’t) with AI coding tools.

Who this is for:

• Software Engineers using or curious about Claude Code and other AI coding assistants
• Tech leads and builders experimenting with AI-first processes

Come prepared:
Bring your questions, your own stories using Claude Code (good, bad, and weird), and be ready for an open conversation about how AI is reshaping software engineering in the real world.

--------
Compound Functional Requirements :

• AI Engineer prompts capture new functional requirements from meeting transcriptions, chats, emails, and integrates them with the existing requirements
• Conflicts/gaps in requirements are resolved before coding all E2E tests derive from functional requirements.
• If requirements change E2E tests are updated/deleted new requirements = new E2E tests
• AI Engineer workflow builds new features/fixes bugs aligned with requirements, and tests E2E tests locally, before merge, for maximum coverage

Picking the right LLM, learning the hard way:

• We test a set of AI models against our own support codebase, not synthetic benchmarks
• A structured rubric and an AI judge score every answer across quality, cost, and deflection rate
• Results are audited when something looks off, including the judge itself
• Cost, quality, and deflection rate tell different stories depending on how you measure
• We leave with a final recommendation and a few things we'd do differently