Fast Thinking Teams for AI Success

Howie Liu believes that in the AI era, companies must fundamentally reinvent themselves by adopting a founder's mindset and getting back into the details. As Airtable's CEO, he's transformed his own role from executive oversight to hands-on building, becoming what he calls an "IC CEO."

This shift stems from Liu's conviction that AI represents a paradigm shift unlike previous technology transitions. With each new model release, novel form factors and UX patterns emerge, making it impossible to understand the solution space without deeply engaging with the technology. As he puts it, "To be continuously relevant and refine product market fit in this era, you have to be in the details. There is no looking at it from 10,000 feet and saying 'we're just gonna throw a bunch of people at this problem.'"

Liu restructured Airtable into two distinct groups: a "fast thinking" team focused on shipping new AI capabilities weekly, and a "slow thinking" team handling more deliberate, infrastructure-focused work. This organizational design allows for both rapid experimentation and thoughtful scaling. He encourages his team to cancel meetings and spend entire days or weeks playing with AI tools, believing that experiential learning is essential for understanding what's possible.

For product teams, Liu sees the boundaries between roles blurring. The most successful people in the AI era are those who can cross traditional role boundaries—PMs who can prototype, engineers with design sensibilities, designers who understand technical constraints. He explains: "As a PM you need to start looking more like a hybrid PM-prototyper who has some good design sensibilities." This polymathic approach enables teams to move faster and build more cohesive products.

Liu personally spends hours daily using AI tools, intentionally becoming Airtable's highest-cost AI user. He creates weekend projects to force himself to use new AI products in meaningful ways, believing that the value-to-cost ratio of AI is "a crazy ratio" when applied smartly to high-value problems.

For leaders navigating this transition, Liu's advice is clear: if you were founding your company today with the same mission, how would you execute using a fully AI-native approach? If you can't leverage your existing assets effectively in this new paradigm, you should consider finding a buyer and starting fresh. As he puts it: "If you really care about this mission, go and start the next incarnation of it."

The practical implication for teams is that everyone must become more versatile and autonomous. Rather than specializing narrowly, people should develop minimum competency across adjacent disciplines while maintaining depth in their core expertise. This "upside-down T" model of skills enables the rapid experimentation and iteration that AI-native execution requires, replacing the deterministic planning and siloed execution that characterized previous eras of software development.

Liu's approach suggests that in the AI era, the most successful companies won't be those that simply add AI features to existing products, but those willing to fundamentally rethink how they operate, breaking down role silos and embracing a more fluid, experimental approach to building.