Sequoia’s Lin Is Optimistic About AI’s Impact

Rosebase: AI for Institutional Memory in Financial Services

Key Concepts:

• Rosebase: An AI platform for asset managers focused on leveraging proprietary data and institutional memory for improved decision-making.
• Institutional Memory: The accumulated knowledge, experience, and data within an organization.
• AI Agents: Autonomous entities powered by AI, designed to understand and reason over data.
• Disparate Data Systems: Multiple, unconnected data sources within an organization (e.g., accounting, trade data, CRM).
• Generative AI: AI models capable of generating new content, like text or images, often used for tasks like creating reports or summaries.
• Observability: The ability to understand the internal state of a system based on its external outputs.

I. Introduction & The Problem Rosebase Solves

The discussion centers around Rosebase, an AI platform designed for asset managers. The core problem Rosebase addresses is the difficulty financial institutions face in effectively utilizing their vast amounts of proprietary data – what they term “institutional memory.” Despite recognizing the inherent value in this data, organizations struggle to access, reconcile, and analyze it efficiently, hindering informed decision-making. The speakers highlight that previous attempts to solve this, even internally at firms like Thrive Capital, relied heavily on foundational technologies like OpenAI, but lacked the specialized infrastructure and depth needed for nuanced financial applications. Mike, the founder of Rosebase, emphasizes that current market products primarily focus on time savings (faster decks, models) while Rosebase prioritizes better decisions through deep data analysis.

II. Rosebase’s Technology & Functionality

Rosebase connects to all of a customer’s data systems – including documents, accounting, trade, position information, and CRM data. Its AI agents are designed to understand the data, identify connections, and resolve inconsistencies. A key example provided involves a large credit originator needing to assess potential trades. Previously, this required manual data export, reconciliation, and cross-referencing with credit documents, a process prone to errors. Rosebase provides a “complete view” by automating this process, reducing the risk of overlooking critical data points. The platform’s ability to handle financial nuances – reconciliation, restatements, EBITDA versions, document versions (adjusted, approved) – is a core differentiator. This is described as an “infrastructure problem,” a “security problem,” and a “product problem” in addition to being an “AI one.”

III. Sequoia’s Investment Thesis & Founder Background

Alfred, a co-steward at Sequoia, explains the firm’s investment in Rosebase. Sequoia’s approach prioritizes strong founders and a compelling market. Michael and Ibo, Rosebase’s founders, met at MIT and possess complementary expertise: Michael from Stripe and RightOcean understands the problem space, while Ibo has finance leadership experience. Alfred stresses the critical need for accuracy in financial decision-making, stating, “People use Rosebase to make better decisions…and they need to make sure the numbers are right.” The investment thesis is based on the belief that Rosebase will empower users to make more profitable decisions. The round is a combined seed/Series A totaling $50 million.

IV. Generative AI, Human Oversight & the Future of Work

The conversation shifts to the broader implications of Generative AI. Avril, also at Sequoia, acknowledges concerns about a “dystopian future” and research (like that from Sattari and Harvard) suggesting AI hasn’t yet surpassed human performance. However, she maintains an optimistic outlook, believing AI will automate routine tasks, freeing humans for more strategic and creative work. She emphasizes the importance of maintaining “a human in the loop” for critical decisions, particularly in areas like retirement, benefits, and insurance. Rosebase is designed with this principle in mind, supporting rather than replacing human judgment.

V. Operational Challenges & Growth Plans

Michael details Rosebase’s immediate operational priorities. Securing data within customer environments is paramount, requiring significant investment in infrastructure and security engineering. The company is also expanding its applied AI research team to enhance the agents’ ability to understand and reason about complex financial data. Rosebase plans to triple or quadruple its team size across its San Francisco and New York locations.

VI. Long-Term Outlook & Legacy Software

The discussion touches on the potential for IPOs of AI companies like Anthropic and OpenAI, but acknowledges a longer-term timeline. Alfred expresses confidence in companies like Snowflake’s ability to adapt to the AI revolution, noting that AI is software and legacy software companies can evolve. He believes Snowflake’s success hinges on its ability to integrate AI into its existing business. He draws a parallel to Oracle, which remains relevant despite being a legacy software provider.

Notable Quotes:

• Alfred: “This is not something you can get wrong. People use Rosebase to make better decisions with AI, and they need to make sure the numbers are right.”
• Avril: “We can think of the world as a dystopian world, but at the same time, we just are very, very optimistic that the impacts are real.”
• Michael: “Agents are only as good as the data they operate on.”
• Alfred: “Rosebase is an AI native company, and so they’re focused on building the way that software is built today.”

Data & Statistics:

• $50,000,000: The amount of Rosebase’s combined seed/Series A funding round.
• Triple/Quadruple: Projected team growth for Rosebase in the coming year.

Conclusion:

Rosebase is positioned as a specialized AI platform addressing a critical need within the financial services industry: unlocking the value of institutional memory. Its focus on deep data understanding, security, and nuanced financial applications differentiates it from broader AI tools. Sequoia’s investment reflects a belief in the founders’ expertise and the potential for Rosebase to significantly improve decision-making in a sector where accuracy is paramount. The conversation highlights the importance of human oversight in conjunction with AI, and the ongoing evolution of software infrastructure in the age of generative AI.