Intent-Driven Delivery

CodeAI like the wind.

For years I coded like the wind. Now I build the systems that let AI do the coding — and prove what it ships is right.

dougfay@ailikethewind.com

I help organizations move from using AI as a coding assistant to running AI as the development organization, with the architecture, verification and cost controls to match.

The method is Intent-Driven Delivery. It is grounded in the AI-native practices Anthropic, OpenAI and Google have described publicly, and proven on a multi-tenant SaaS platform where all net-new code is AI-authored under mandatory human verification, running on a continuous autonomous build loop I designed and operate.

Engineer the system around the AI, not just the code it produces.

  • Right-sizing architecture to the company

    The best design is the one a team can afford to run and staff today.

  • Designing for AI-first development and testing

    Near-total AI authorship changes how work is specified, how tests are trusted, and where humans must stay in the loop.

  • Multi-agent concurrency

    Several agents on one codebase, with work queues, ownership boundaries and merge discipline so they don't overwrite each other.

  • The autonomy ladder

    Approve every step → approve the plan → approve only the result. Each step needs a stronger automated gate underneath before you climb it.

  • Usage efficiency as architecture

    Scheduled routines, continuous loops and direct API calls have very different economics. Context engineering, execution model and budget guardrails decide which method is affordable.

  • Verification and auditability

    When AI writes the code and the tests, proving correctness becomes the job. Test ratios, written verification records, infrastructure as reviewed code and keyless deploy credentials are what a security team signs off on.

Who you would be working with

I'm Doug Fay. I started as a software engineer at EDS, and in 2000 I was lead application architect for the launch of Fandango.com. That was the code-like-the-wind era.

I spent the next 25 years at Cinemark, going from senior engineer to director of application development to Chief Technology Officer, a role I held for nearly two decades. My teams built and ran the systems a national theatre chain depends on every night: point of sale, gift cards, the Movie Club subscription, Cinemark.com and the mobile app, and the ticketing, payments, fraud and loyalty platforms behind them. I also established the company's cybersecurity governance and helped build its AI governance framework.

So I care less about the demo and more about whether a system can be operated, secured, audited and paid for — at scale, on a Friday night, when it matters.

1994Software engineer, EDS Systems Engineering Development Program
2000Lead application architect, Fandango.com launch
2001–2026Cinemark — senior engineer, director of application development, then CTO from 2006
2026Founded AI Like the Wind, Austin, Texas