The Eightfold co-founders just cracked one of the workplace's most expensive problems: what happens when crucial teammates are unreachable. Their new startup Viven emerged from stealth today with $35 million in seed funding to build AI digital twins that let employees query unavailable colleagues' knowledge instantly.
Viven just solved the $100 billion workplace coordination problem that every company faces but nobody talks about. When your colleague with critical project info is on vacation in Bali or sleeping in a different time zone, your entire team grinds to a halt. It's the kind of friction that makes enterprise software executives wake up in cold sweats.
Ashutosh Garg and Varun Kacholia know this pain intimately. The Eightfold co-founders built their AI recruiting platform into a $2.1 billion unicorn by automating human bottlenecks, and now they're applying that same playbook to workplace communication. Today, their stealth startup emerged with $35 million in seed funding from Khosla Ventures, Foundation Capital, and FPV Ventures.
The concept sounds like science fiction but works remarkably simply. Viven creates a specialized large language model for each employee by ingesting their internal communications - emails, Slack threads, Google Docs, project files. Other team members can then query that person's "digital twin" as if they're having a conversation with the actual colleague. "When each and every person has a digital twin, you can just talk to their twin as if you're talking to that person and get the response," Ashutosh Garg told TechCrunch.
The privacy implications are staggering, but that's exactly where Viven thinks it has cracked the code. The company's "pairwise context and privacy" technology determines precisely what information each digital twin can share with specific colleagues. Personal conversations about weekend plans stay locked away, while project updates and technical knowledge flow freely to authorized team members. Every query gets logged in a transparent history that the original employee can review - creating a natural deterrent against inappropriate snooping.
"It's a very hard problem to solve, and until recently, it was unsolvable," Foundation Capital general partner Ashu Garg told TechCrunch. The breakthrough came from combining advances in large language models with sophisticated privacy controls that didn't exist even two years ago.
