The journey to enhance spreadsheets through artificial intelligence continues with the emergence of a new player, Meridian. This innovative company has recently unveiled a more integrated IDE-based approach to financial modeling, backed by substantial funding. On Wednesday, Meridian announced it has raised $17 million in seed funding, achieving a post-money valuation of $100 million.
John Ling, CEO and co-founder, shared with TechCrunch, "We aim to make financial modeling and spreadsheets significantly more predictable and auditable. How can we transform a process that typically takes hours into a mere 10 minutes?"
The funding round was spearheaded by Andreessen Horowitz and the General Partnership, with contributions from QED Investors, FPV Ventures, and Litquidity Ventures. Currently, Meridian is collaborating with teams at Decagon and OffDeal, having secured $5 million in contracts just in December.
AI startups have increasingly targeted Excel agents, motivated by the high costs associated with human-led financial analysis. Unlike previous Excel agents, such as Shortcut AI, which integrated directly into Excel, Meridian operates as an independent workspace, similar to Cursor. This design allows the application to function like an IDE, seamlessly incorporating various data sources and external references that might typically create obstacles.
Located in New York, the Meridian team comprises experts from AI companies like Scale AI and Anthropic, alongside seasoned professionals from financial institutions such as Goldman Sachs.
Ling highlights a significant challenge for Meridian: balancing the stringent demands of financial clients with the inherently unpredictable nature of AI models. "If you approach ten different software engineers at Google for a new app feature, you'll likely receive ten distinct implementations, which is perfectly acceptable," Ling explains. "However, if you consult ten banking analysts at Goldman Sachs for valuation models, you would probably end up with ten nearly identical workbooks."
To address this, the Meridian team has focused on enhancing the auditability and determinism of their outputs while preserving the adaptability of LLM-based tools. The outcome is a blend of agentic AI and traditional tools, effectively reducing the inaccuracies that can hinder many enterprise applications.
Li, another key team member, emphasizes, "Our mission is to eliminate uncertainty right from the LLM process. You'll have a clear understanding of how the logic operates, and all assumptions that feed into the model will be transparent."