Gabe Stengel - Building Investing Superintelligence - [Invest Like the Best, EP.492]

Summary

Gabe Stengel, co-founder and CEO of Rogo, makes the case that the most important shift in finance is not raw model capability but the infrastructure that makes AI usable inside regulated workflows. He describes Rogo’s evolution alongside model jumps, from early multi-call “Rube Goldberg” systems to simpler architectures as frontier models improved, and argues the moat is the harness: data plumbing, evals, routing, compliance, and auditability. The company’s current wedge is private-market dealmakers and bankers, where it helps with data rooms, diligence, deck markups, CRM updates, and LP reporting. Stengel believes finance software will move from seat-based pricing toward usage and eventually outcome-based pricing, and that AI will compress transaction timelines, raise liquidity, and unlock new capital-raising and M&A workflows. He is equally focused on company-building: aggressive hiring, internal AI adoption incentives, and an internal “company brain” called Shrek that captures knowledge and helps employees act faster.

Notes

• Rogo’s product roadmap tracks frontier model eras closely, with capability jumps changing what workflows are feasible.
• Early Rogo was a 60-call orchestration stack, not a simple single-model app.
• Model routing and evals matter because cost, latency, and quality all have to be traded off.
• The core user today is a transaction-oriented finance professional, not yet a broad public-markets hedge fund buyer.
• Rogo supports end-to-end deal workflows: data rooms, diligence, coordination, and closing.
• A concrete use case is MD deck markup review over email, with turnaround cut from days to minutes.
• The firm is building around regulated-workflow requirements, not just model quality.
• Auditability is a product requirement because users need to inspect data lineage, assumptions, and reasoning.
• Copilot-to-autopilot AI in finance requires debuggability and explainability.
• The company’s moat is the surrounding plumbing: compliance, evaluation, lineage, and integrations.
• Rogo’s long-run wedge is finance’s messy, niche, regulated data landscape.
• Rogo says it can reach $5 billion revenue by going deep across finance niches.
• The toughest open problem is memory compaction across long-running agent relationships.
• The hard part of AI is the harness around intelligence: storage, retrieval, triggering, and persistence.
• Rogo is sold like classic finance software today, per seat.
• The buyer mental model is Bloomberg, FactSet, Capital IQ, and PitchBook.
• Stengel wants finance software to evolve from seats to usage and then to outcomes.
• The roadmap is still very early, by Stengel’s own estimate.
• AI could compress capital-markets timelines from months to minutes in some workflows.
• Rogo’s current product philosophy favors transparency over opacity in regulated settings.
• Internal AI adoption is managed like an operating KPI, with monthly reporting and stack rankings.
• Rogo’s internal memory system records all conversations and filters them into a company brain.
• The company brain exists to reuse knowledge across teams and geographies.
• Shrek is the name of Rogo’s internal dashboarded company brain.
• Shrek is both reactive and proactive, including meeting prep tied to client use cases and ROI.
• Stengel pitches Rogo talent on applied AI at the point where AI intersects with humans.
• He frames finance as upstream of broader economic progress through capital allocation.
• Stengel thinks AI could make capital markets vastly more efficient and even support a $500 billion company outcome.
• He argues faster capital access would help entrepreneurs, small businesses, and emerging markets.
• Private markets may standardize and become more liquid as AI reduces unstructured-data friction.
• Human trust may remain important in some small-business and long-tail M&A situations.
• AI-native finance means constantly reinventing the business with no sacred cows.
• The company needs aggressive hiring because finance buyers will make AI decisions soon regardless.
• Stengel embraces risk if it increases the chance of a massive outcome.
• He thinks finance investors underestimated the market because Silicon Valley lacked intuition for it.
• Rogo’s first iterations were too weak, causing skeptical investors to dismiss the category.
• Distribution in finance is a trust problem as much as a product problem.
• Rogo has acquired six fledgling financial AI startups.
• Former founders with product intuition are especially valuable hires.
• Private finance incumbents have not yet experienced a true AI innovator’s dilemma shock.
• Enterprise value may shift from people into software, data, and systems.
• Firms should map the deal lifecycle to find where they truly own unique data or context.
• He imagines digital clones of top bankers to transfer context and relationships across the org.
• The biggest uncertainty is how much relationship alpha survives in private markets.
• He is direct about fundraising, but says repeated exposure builds conviction over time.
• Founders can be scared and still project conviction.
• Stengel is motivated by paranoia that momentum could stall.
• Winners in AI finance may become black holes for talent, capital, and brand.
• The right team is expensive, and laggards risk being outcompeted by better-resourced AI firms.