Engineering in the AI era
The rules are changing. Story sizing, feedback loops, team structure, architecture, and how we verify business value — all of it needs to evolve. This methodology is based on building AI-first products and directing agentic AI systems for Fortune 500 companies.
Software engineering practices were designed for a world where building was expensive and slow. Every methodology — Agile, Scrum, SAFe, Kanban — was created to manage the constraints of human-speed development. Story points exist because we needed to estimate how long human work takes. Sprint reviews exist because we could only show progress every two weeks.
AI is dissolving these constraints. When a task that took a sprint takes a day, the entire planning and estimation model breaks. When you can prototype three approaches in the time it used to take to debate one, the decision-making process needs to change. When development is 5-10x faster, the bottleneck shifts from building to validating.
Engineering practices
Story sizing when AI compresses timelines
The old estimation models assumed human-speed development. That assumption is now false.
Feedback loop compression
When you can build in hours instead of weeks, waiting two weeks for a sprint review is absurd.
Big bet engineering
When building is cheap, you can try things that used to be too expensive to attempt.
Architecture for AI velocity
Designing systems that AI can work on effectively. Not all codebases are equal.
UX testing at speed
When you can build a prototype in hours, you can test it with users the same day.
Organization
Business verification during development
The old model: build it, then verify. The new model: verify continuously, build iteratively.
Team structure for AI-augmented development
Fewer people, broader scope, more autonomy.
Incentive realignment
Every organizational dysfunction has an incentive structure behind it.
The execution framework
Knowing what's wrong is the easy part. Actually fixing it is where most transformations die.
Ready to modernize your engineering practices?
These practices are tested in real organizations. Let's talk about how to apply them to yours.
Start a Conversation