The archive · Money & Fintech · Product decision · 2022-2026
Rogo's AI-banker bet: from a Princeton thesis to $2B automating juniors' grunt work
Three ex-bankers bet finance-native AI agents could automate junior bankers' grunt work - 35,000+ users, $160M Series D at $2B
Rogo
What the business is
Agentic AI platform for investment banking and finance: automates research, Excel modeling, deal screening, CIM generation and due-diligence workflows, integrating banks' proprietary data with market databases and filings.
Starting capital：$50M Series B (Apr 2025, total $75M); $160M Series D (Apr 2026, led by Kleiner Perkins, valuation ~$2B, total over $300M)
How it started
Founded in January 2022 by Princeton classmates Gabriel Stengel, John Willett and Tumas Rackaitis, who had worked at J.P. Morgan and Lazard. The release of GPT-3 for developers let them expand their senior thesis - a chatbot for econometrics - into a working prototype, and they left their finance jobs to build it.
What happened
Raised a $50M Series B in April 2025 (total $75M) to invest in financial reasoning models and autonomous agents. The platform automated analyst-level work - research, valuation analysis, Excel models, presentations - for investment banks, asset managers and private equity. By April 2026 it supported 35,000+ professionals at 250+ institutions, acquired UK-based Plux AI and the AI agent startup Offset, launched Felix (an end-to-end agent that handles deal screening, CIM generation, buyer outreach and data-room diligence), and closed a $160M Series D led by Kleiner Perkins with Sequoia, Thrive Capital, Khosla Ventures and J.P. Morgan Growth Equity participating.
How it ended up
Still scaling: the April 2026 Series D valued Rogo at roughly $2B, and the company is expanding globally, deepening bank partnerships and turning Felix into an end-to-end workflow system while banks debate what automation means for entry-level hiring.
Background
Rogo was founded in January 2022 by three Princeton classmates - Gabriel Stengel, John Willett and Tumas Rackaitis - who had worked at J.P. Morgan and Lazard. When GPT-3's API opened to developers, they saw a way to turn their senior thesis, a chatbot for econometrics, into a real product, and left finance to build an AI platform for the industry they knew best.
The bet was that the repetitive work that consumes junior bankers - research memos, Excel models, pitch decks, due diligence - could be automated by finance-native agents, and that banks would trust those agents with client-critical workflows. Rogo integrated proprietary deal data, market databases, regulatory filings and CRM systems, then embedded experienced bankers inside client institutions to drive adoption from analysts up to managing directors.
A $50M Series B in April 2025 brought total funding to $75M. By April 2026 the platform supported 35,000+ financial professionals at 250+ institutions including Rothschild & Co, Jefferies, Lazard, Moelis and Nomura; the company had acquired UK-based Plux AI and the agent startup Offset, launched Felix (an end-to-end agent for deal screening, CIM generation, buyer outreach and data-room diligence), and closed a $160M Series D led by Kleiner Perkins at roughly a $2B valuation.
Rogo's branded-search momentum tracked the funding: it climbed 2,014 positions month-over-month in Analyze AI's ranking of fastest-growing AI companies as of early 2026, with $20M+ in reported ARR. The case shows a vertical bet succeeding by owning the workflow - data integrations, embedded bankers and domain-specific agents - rather than competing on model quality alone.
What has to be true
- Junior banking work is structured, high-stakes and repetitive - a perfect first wedge for automation that generates visible ROI for banks.
- Founders who had done the job themselves knew exactly which workflows to target and how banks would buy.
- Integrating proprietary firm data and embedding forward-deployed bankers made Rogo part of the bank's operating process, raising switching costs.
- Client concentration in a few hundred institutions means each enterprise deal is large, and the category's growth shows in both revenue and branded search.
What can be applied
The wedge is the workflow, not the model: training finance-native agents on proprietary data and embedding them in banks' own systems made Rogo hard to displace even as general-purpose AI improved.
Aftermath
As of May 2026, Rogo was valued at roughly $2B after its April Series D, supported 35,000+ finance professionals at 250+ institutions, and was expanding its agentic platform (Felix) and global footprint while the banking industry debated how much AI adoption will reshape entry-level roles.
Sources
- Kleiner Perkins leads Rogo's $160M raise to build the AI operating system for investment banking
- Rogo Raises $50 Million to Develop AI-Powered Investment Banker
- Can AI Outperform Junior Bankers? What Rogo's $2 Billion Valuation Means
- The 50 Fastest-Growing AI Companies in 2026
spotted an error? The archive wants to know.
Your turn
You just read one. Describe what you are building, and see who is betting on the same thing.
Free account · 3 free questions · no card