Key Takeaways

  • In May, Cloudflare executed the first corporate restructuring in its history to eliminate handoffs between teams and reorganize around automated execution.
  • The company built Cloudflare OS, an internal governance and application platform designed to let employees query data and run AI models without exposing sensitive customer records.
  • Go-to-market and general administrative groups use Cloudflare OS more than software engineering teams.
  • Cloudflare collapsed a three-step customer email pipeline (business intelligence, marketing copy, and marketing operations) into a single workflow run by one product manager.
  • Dane Knecht projects that enterprise token spend over the next 18 months will concentrate heavily on security defenses against automated threat actors.

Cloudflare Replaced Multi-Team Queues With One Platform

In May, Cloudflare did something it had never done before: it executed a company-wide restructuring. The goal was not cost-cutting or typical enterprise streamlining. Chief Technology Officer Dane Knecht saw how machine learning was changing internal production lines. As Knecht put it, “We kind of thought of it as like a factory floor where you needed to actually rebuild rethink how things happen.”

The problem was handoffs. In growing tech companies, simple operational tasks get chopped into tickets spread across specialized departments. When a product manager wanted to contact customers, they ran into a wall of queues.

“The way a product manager would send an email to customers, it involved three teams,” Knecht said. “You went to BI to go get a cohort of the people that you wanted to send an email to. You went to the marketing team to help you design the email and get the voice right. You went to the marketing operations team to get the email out.”

Each hop added days of delay, review meetings, and friction. Speeding up each department individually solved nothing because the coordination tax remained identical.

Non-Technical Staff Run Most Internal AI Workflows

To break that bottleneck, Cloudflare built Cloudflare OS. The platform originated inside their security group because the company wanted strict controls against proprietary customer records leaking to external AI models. As Knecht explained: “Cloudflare OS was our internal platform for how our employees use AI every day. It actually goes back to security roots where we were very worried about customer data internal data leaking out.”

Once the platform established secure boundaries around company databases, usage shifted in an unexpected direction. Engineering was not the primary consumer. Knecht noted that “it's actually used more by our go to market teams and our G&A teams than it is even used by our engineering teams.”

Instead of coordinating with BI, marketing design, and marketing operations, a product manager now opens Cloudflare OS and types in the criteria for the customer cohort in plain English. The platform pulls the exact recipient list from company data stores, reviews the product manager's draft, and rewrites the copy to match Cloudflare's editorial voice. From there, the system routes the message directly to the manager for a one-click review.

“We didn't go from three teams getting faster at doing that,” Knecht said. “We went to one tool that can let a product manager now have a superpower.”

This shift changes how startups should think about headcount. When software handles database querying, voice consistency, and operational delivery, you do not need horizontal coordination layers. You need autonomous operators equipped with secure internal access.

What to Do With This

Map the single most frequent inter-departmental ticket queue in your company this week, like data pulls or copy approvals. Build an internal script or LLM pipeline connected to your read-only database that lets the originating employee run the process and request approval with one click.