How I build, think, automate, and ship in the AI era

💡 My Philosophy

• AI tools are the new Stack Overflow: accelerators for learning, building, and shipping.

• They allow engineers to become 10× more autonomous, but only with intentional practice.

• Using AI effectively is like mastering a new programming paradigm:

  • clear specifications (SDD — Spec-Driven Development)
  • strong tests and guardrails (TDD — Test-Driven Development)
  • structuring and managing context (/clear /compact)
  • prompting with intent,
  • and designing systems that AI can understand and extend.

• I believe we’re in the AI era of building software where engineers focus on clarity of architecture, clean boundaries, and strong tests, while GenAI handles 90% of code generation.

• My Agentic Coding workflow reflects that shift:

  1. Explore → 2. Plan (SDD) → 3. Write tests (TDD) → 4. Generate code → 5. Verify → 6. Review

⚡ Context

AI is now central to how I prototype, iterate, and build product, especially in high-trust domains like DeFi, compliance, payments, and asset management.

I use AI to:

• ship ideas faster,
• automate workflows,
• augment complex reasoning,
• and build powerful new capabilities with small teams.

At Crystal Intelligence I drove AI adoption across engineering. Today I build agentic systems for clients through Waver Labs, including a forward-deployed engagement with Rubbr on coding-agent tooling.

🛠 Professional Use

• Led AI adoption for 40+ engineers at Crystal Intelligence, structured in three phases: pilot → ongoing → steady