Zaro.ai, a London-based enterprise AI startup, has emerged from stealth with $5.1 million in pre-seed funding led by Cherry Ventures. The company was founded by Michael Bajwa and Qian Zheng, who previously built AI agent products at Convergence before it was acquired by Salesforce, where the team worked on Agentforce. Additional investors include Thomas Wolf of Hugging Face, Thomas Dohmke of GitHub, Charlie Songhurst, and Convergence co-founders Marvin Purtorab and Andy Toulis.
Zaro’s platform is designed to address a fragmentation problem the founders observed firsthand: enterprise AI deployments typically scatter context, workflows, and outputs across multiple disconnected tools, preventing intelligence from compounding over time. The company’s answer is a single workspace where company data, AI agents, and custom applications operate within one unified context layer owned by the business rather than the software vendor.
The platform is model-agnostic, routing simpler tasks to lower-cost AI models and reserving frontier models for more complex work, a approach Zaro claims can reduce costs by roughly ten times compared with frontier-only deployments.
Cherry Ventures partner Dinika Mahtani backed the company before the product was built, arguing that Zaro is the first platform where AI demonstrably gets smarter the longer it operates inside an organisation. Zaro currently has eight employees, five of whom built agents together at Convergence prior to the Salesforce acquisition.
