In partnership with
Most of the industry is still waiting for AI agents to be "ready" before trusting them with anything real. Ritesh Patel isn't waiting. He's already running live events, logistics, contracts, ops, safety, on agents he built himself. The lesson for project delivery isn't the tech. It's that the people closest to the work are the ones who make agents actually work.
🎧 Hit play on Ritesh Patel — AI Agents Are Running Entire Businesses, Is Project Delivery Next? (Ep 122, 43:05). Worth the walk to work.
Ritesh Patel co-founded The Ticket Fairy — a Y Combinator-backed ticketing and event-marketing platform and has spent fifteen years in the unglamorous end of live events: the load-ins, the vendor contracts, the 2am problem nobody planned for. That background is the whole point. He argues the reason most AI agent projects stall isn't the model, it's that they're built by people who've never actually done the job the agent is meant to do.
His answer is a custom AI harness: not one giant do-everything assistant, but a set of narrow, industry-specific agents wired to the real production workflow, logistics, vendor management, contracts, continuous feedback. They don't just react; the interesting ones are proactive, flagging operational and safety issues before a human would spot them. The value shows up because the person who scoped them knew exactly where the work goes wrong.
The direct read-across for project delivery is uncomfortable. Everyone wants the platform that runs the project. Ritesh's experience says you get there by codifying the ground-level knowledge first, in small pieces, from people who've carried the risk, not by buying a transformation programme and hoping the domain expertise turns up later. He's candid on jobs, too: agents absorb the repetitive coordination, and the roles that survive are the ones anchored in judgement and the parts of live events that only work in person.
He's not starry-eyed about it. The back half gets into recursive self-improvement, where the efficiency gains actually land, and the societal bill that comes with automating coordination at scale. His 2031 prediction for live events is worth the final few minutes on its own.
Why it matters: AI agents aren't a someday capability waiting on a better model — they already run real operations today. The winners aren't the ones with the best tech; they're the ones who let the people who do the work design what the agents do.
👉 One thing to do this week: pick the most repetitive coordination task on your project — the chasing, the status-gathering, the vendor back-and-forth — and write down every step the way the person who actually does it would. That description, not the model, is the hard part of an agent.
Links and Stuff
Ritesh Patel (Co-Founder & Co-CEO, The Ticket Fairy) — LinkedIn
The Ticket Fairy — theticketfairy.com
Listen on
Subscribe to the Project Flux newsletter — the week's hottest AI and project-delivery news in your inbox every Monday.
Write docs 4x faster. Without hating every second.
Nobody became a developer to write documentation. But the docs still need to get written — PRDs, README updates, architecture decisions, onboarding guides.
Wispr Flow lets you talk through it instead. Speak naturally about what the code does, how it works, and why you built it that way. Flow formats everything into clean, professional text you can paste into Notion, Confluence, or GitHub.
Used by engineering teams at OpenAI, Vercel, and Clay. 89% of messages sent with zero edits. Works system-wide on Mac, Windows, and iPhone.
Project Flux is independently owned, funded, and operated by its founder. Any external references or features are for community benefit only and do not imply endorsement, control, or ownership.