Vorflux raises $15M seed for AI coding 'autopilot'
What's the deal? Rippling cofounder Prasanna Sankar has unveiled Vorflux, a startup building an AI "autopilot" for software engineering, with $15 million in seed funding. Y Combinator, Peak XV Partners, PowersetDealroom has a profile for this one. Try Dealroom →, and AllianceDealroom has a profile for this one. Try Dealroom → backed the round, alongside angel investors Parker ConradDealroom has a profile for this one. Try Dealroom →, Immad AkhundDealroom has a profile for this one. Try Dealroom →, and Balaji Srinivasan.
What's the endgame? Vorflux aims to automate the entire software development cycle rather than just code generation. It sets up development environments, builds implementation plans, coordinates code changes, runs tests, reviews code, and manages merges.
How it works: Rather than train its own models, Vorflux uses models from different AI labs and lets companies apply their own engineering workflows to each session. That layer sits inside what Sankar calls a "harness" — a way to capture a team's practices and standards.
Why now? Sankar argues AI models have advanced far enough to do more than generate code. "The models quietly got good enough to fly the whole route, but the tools never caught up. So we built the autopilot," he said in a post on X.
The gap: Cloud-based coding assistants can write code but cannot run an application's full environment, Sankar said, forcing developers back to their own machines to test and validate changes. He frames the shift as engineers moving toward defining the judgment and principles AI systems should follow.
Sankar, who remains chief technology officer at Rippling, calls Vorflux his "high-octane Ferrari."
The signal: "Writing code became a commodity. Bottlenecks moved," Sankar said. Vorflux reflects a broader push to reposition AI coding tools from assistants that need constant human approval toward systems that run development end to end.
Read more: The Economic Times
