Coffee with a Founder Podcast: James Everingham, Guild AI

29- Sep-2026

On this episode of Coffee with a Founder, Beck Bamberger sits down with James Everingham, serial entrepreneur and CEO of Guild AI, to talk about what it takes to build for a world of AI agents.

Drawing on decades in tech, including time at Meta, James explains why a strong observation is a better starting point for a company than the desire to start one. He shares how Guild AI is helping enterprises address the less visible challenges of AI agents: tracking their actions, managing costs, and establishing who is accountable when an agent acts.

James also reflects on the cycles of founding companies, what failure can teach you, and why founders need to keep questioning their assumptions as technology changes. It’s a candid conversation about building something useful, learning quickly, and finding some joy in the process.

show notes:

00:00 - Building Guild's Production Layer

06:17 - Motivations for Starting a Company

08:38 - Changing perspectives on experience

12:24 - Using multiple software models

15:26 - Setting ambitious revenue goals

Read More About The Episode:

00:00 — Building Guild AI’s production layer for agents

James Everingham, founder and CEO of Guild AI, explains what enterprises need to put AI agents to work safely. Guild AI is building tools to monitor agent behavior, control access, track costs, and make actions auditable. James draws on his experience at Meta to explain why these controls become harder to build as agent use grows.

06:17 — What motivates James to start a company

James reflects on his work across startups and large companies. For him, the reason to start a company is a clear observation about a problem that needs solving. He also talks about the appeal of building, learning, and beginning again when an idea does not work.

08:38 — How experience can change your perspective

Experience helps founders recognize patterns, but James says it can also leave them relying on assumptions that no longer hold. He describes reconsidering his belief that producing high-quality software must take more time as new tools change how quickly teams can build.

12:24 — Why Guild AI uses multiple models

James discusses the risks of tying an enterprise system to a single AI model while the technology is changing so quickly. Guild AI’s infrastructure can use different models, giving teams flexibility to choose what works for each task.

15:26 — Setting ambitious goals

James shares a thought exercise he uses with teams: ask what would have to change to achieve a result ten times greater than the current goal. He recalls applying that thinking at Meta to rethink how quickly software could move into production.

17:15 — Advice for founders

James closes with a simple test for starting a company: find a problem so clear and necessary that you feel compelled to solve it. He encourages founders to question their assumptions, learn through setbacks, and enjoy the work along the way.