Zeroset, a San Francisco-based artificial intelligence startup developing enterprise memory and state infrastructure, announced a $5.2 million pre-seed funding round on October 6, 2026.
The round was co-led by Gradient and 2048 Ventures, with participation from Leblon Capital and other investors. The funding will support engineering recruitment, model training, product development and early enterprise deployments.
Founded by Akshat Kannan and William Zhang, Zeroset is building technology that enables AI agents to understand how organizations operate, including their internal processes, decisions and changing operational conditions.
The company’s first product, Nebula, is designed to provide AI agents with persistent access to enterprise workflows and operational context. Rather than relying solely on document retrieval, the platform aims to represent how work progresses across business systems and how organizational information changes over time.
Nebula connects with enterprise applications including Microsoft 365, Outlook, Teams, SharePoint and GitHub. It captures information generated through employee and AI-agent activities, including messages, approvals, task handoffs and decisions, to build structured representations of business operations.
The platform is intended to help AI agents retrieve relevant operational context, identify dependencies between tasks and understand the current state of ongoing workflows. This approach is designed to reduce the need for agents to reconstruct company-specific processes from disconnected data sources.
Zeroset is also developing enterprise-specific models that can learn from recorded operational activity and changes in business processes.
Nebula was in a closed research preview with selected design partners at the time of the funding announcement. The company plans to use its new capital to expand technical development and support initial deployments with enterprise customers.
