Meridian, a New York-based startup, has emerged from stealth with $17 million in seed funding and a $100 million post-money valuation, positioning itself at the forefront of what it calls “agentic financial modelling”.
The round was led by Andreessen Horowitz and The General Partnership, with participation from QED Investors, FPV Ventures and Liquidity Ventures, a roster that underscores growing investor appetite for AI-driven tools targeting high-value enterprise workflows.
At the heart of Meridian’s proposition is a fundamental question, that can artificial intelligence make financial modelling faster, more predictable and fully auditable?
“Our goal is to make financial modeling and spreadsheets way more predictable and auditable,” CEO and co-founder John Ling told TechCrunch. “How can you take a process that traditionally might have taken several hours and condense it down into like 10 minutes?”
What You Need to Know
AI-powered “Excel agents” have become an increasingly crowded category. The promise is straightforward: automate complex spreadsheet tasks traditionally performed by investment bankers, analysts and corporate finance teams.
However, many existing startups have opted to build directly inside Excel. Meridian has taken a different route.
Rather than functioning as a plug-in, Meridian operates as a standalone workspace, structured more like a software development environment, comparable to coding platforms such as Cursor. This Integrated Development Environment (IDE)-style approach allows Meridian to pull in external data sources, reference documents and financial logic layers in a controlled workspace, reducing friction and improving traceability.
The distinction is subtle but strategic. Excel-based agents often struggle with version control, external dependencies and output verification. By creating a dedicated modelling environment, Meridian aims to give financial teams a clearer audit trail and more deterministic outputs, two features that are non-negotiable in high-stakes finance.
The Determinism Challenge
One of the most significant hurdles facing AI in finance is the inherent unpredictability of large language models.
In software engineering, variation can be tolerated. Different engineers may implement the same feature in different ways without materially affecting outcomes. In banking, however, consistency is paramount.
“If you go to 10 different software engineers at Google, and you want to add some new feature into an app, you’ll probably get like 10 completely different implementations. And that’s totally fine,” Ling says. “But if you go to 10 banking analysts at Goldman Sachs and you ask for 10 valuation models for a company, you would probably get 10 almost identical workbooks.”
This expectation of near-identical outputs clashes with the non-deterministic nature of generative AI systems, which can produce varied responses to the same prompt. Meridian’s technical architecture therefore blends agentic AI with more conventional rule-based tooling, designed to minimise hallucinations and enforce logical consistency.
In practice, this means building guardrails that preserve flexibility while ensuring every formula, data input and modelling assumption can be traced and verified, a prerequisite for regulatory compliance and investor scrutiny.
Early Traction and Revenue Signals
Despite emerging publicly only now, Meridian claims early commercial traction. The company says it is working with teams at Decagon and OffDeal and signed $5 million worth of contracts in December alone, a notable figure for a seed-stage firm.
The demand reflects the high cost of human-led financial analysis. Investment banking models can take hours, sometimes days to construct and review. Automating even part of that process could significantly reduce labour costs while accelerating deal cycles.
Investors appear to see the opportunity. The $17 million seed round places Meridian among the better-funded entrants in the AI-for-finance space, and the $100 million valuation signals confidence in its ability to capture enterprise adoption.
Meridian’s team blends AI research expertise with financial pedigree. The company includes alumni from Scale AI and Anthropic alongside veterans from Goldman Sachs, a combination designed to bridge the gap between technical capability and real-world financial requirements.
The hybrid background may prove critical. Financial modelling is not merely a computational task; it is shaped by industry conventions, regulatory norms and deeply ingrained professional standards. Translating those nuances into an AI system requires domain fluency as much as engineering sophistication.
By focusing on predictability, auditability and deterministic outputs, experts say Meridian is targeting precisely the friction points that have slowed enterprise AI deployment in regulated industries.
Talking Points
It is telling that Meridian is not simply building another AI assistant for Excel, but rethinking financial modelling as an IDE-driven workflow. That structural shift could prove more consequential than incremental spreadsheet automation.
The core tension Meridian is addressing is determinism. In finance, outputs must be predictable, auditable and repeatable, a direct contrast to the probabilistic nature of large language models.
At Techparley, we see this as a maturity test for enterprise AI. It is no longer enough for AI tools to be fast; they must be verifiable. Financial institutions cannot operate on outputs that “look right” — they must be able to trace every assumption and formula.
The decision to operate outside Excel as a standalone workspace may offer Meridian tighter control over logic, versioning and external data integration. That could give it an edge over plug-in based competitors that remain constrained by spreadsheet architecture.
Meridian’s blend of Wall Street veterans and AI lab alumni reflects a broader trend: the most credible enterprise AI companies are increasingly built at the intersection of domain expertise and frontier research.
If Meridian succeeds, it could shift financial modelling from manual craftsmanship to supervised AI orchestration, fundamentally changing how deals are analysed, priced and executed.
——————-
Bookmark Techparley.com for the most insightful technology news from the African continent.
Follow us on Twitter @Techparleynews, on Facebook at Techparley Africa, on LinkedIn at Techparley Africa, or on Instagram at Techparleynews.
