The Finance Startup Bringing Agentic AI to Wall Street

YouTube Video Transcript Summary: Model ML Founders Arie and Chaz Englander

Key Concepts: AI workspace for financial services, agentic systems, cognitive architecture, data integration, automation, perseverance, hiring practices, model improvements, sales strategy, trust-building.

1. Introduction and Background

• Arie and Chaz Englander, founders of Model ML (Winter 24 YC batch), are interviewed.
• They previously founded and sold two other YC companies: Fancy and Fat Llama.
• Model ML is described as an "AI workspace for financial services."

2. Model ML: AI Workspace for Financial Services

• Core Functionality: Model ML is presented as an AI-powered office suite (Word, PowerPoint, Excel) built on an agentic system.
• Agentic System: Mimics a human's access to data and tools within a financial firm.
• Cognitive Architecture: A "brain" that mirrors digital access, including file systems, emails, CRM, data vendors, real-time public information, public filings, and internal data sets.
• User Interface: Overlaid on the cognitive architecture to provide a user-friendly experience.
• Value Proposition: Reduces time spent gathering information by connecting directly to data sources within an Excel-like environment.

3. Current Traction and Market Shift

• Rapid Growth: Chaz states, "In the last seven days, we've signed the same number of contracts as we signed in the whole of Q4."
• Market Turning Point: There's a shift in the sector towards recognizing the "clear tangible value being driven by these products."
• Shift from Testing to Usage: The industry has moved from testing AI solutions to actively using them.

4. Problem Solved and Origin Story

• Previous Workflow: Before Model ML, financial professionals relied heavily on the office suite and Outlook, leading to manual and repetitive tasks.
• Example: Organizing logos in PowerPoint presentations, which is not an efficient use of highly educated analysts' time.
• Origin: Arie and Chaz initially built a genensic system for their own investment activities after selling their second company.
• Automated One-Pager: The system would automatically generate a one-page summary of investment opportunities, pulling data from LinkedIn, Crunchbase, S&P, and review websites.
• Transition to Product: The tool's potential was recognized, leading to its development into a product for top financial firms.

5. Market Penetration and Customer Base

• Significant Adoption: Model ML is used by approximately 10% of the largest private equity firms and investment banks globally.
• Diverse Clientele: Customers include asset managers, sovereign wealth funds, and venture firms.

6. Use Case: Automating Earnings Summaries

• Traditional Process: Analysts spend days creating earnings summaries (slides) by manually extracting data from filings and sources like FactSet.
• Model ML Solution: Model ML connects to data sources, allowing for automated generation of earnings summaries.
• Workflow: An Excel spreadsheet connected to data sources can export data into designs, automatically generating a cover page, main page, and legal page in SharePoint or Google Drive.
• Efficiency: Reduces the time to create these summaries, achieving 90-95% completion automatically.
• Accuracy: Model ML can be more accurate than humans due to its ability to pull data from multiple sources.

7. AI Model Advancements

• Continuous Improvement: AI models have improved significantly, especially since January.
• Function Calling: Improvements in function calling have enhanced agentic systems.
• Vision Models: Vision models have revolutionized file analysis by combining OCR with the ability to read tables and charts.
• Accuracy vs. Humans: Models are becoming more accurate than humans in data gathering and presentation tasks.

8. Market Adoption of AI in Finance

• Changing Landscape: Investment banks and private equity funds are now actively buying software, driven by AI curiosity.
• Shift in Mindset: The shift from proof-of-concept projects to actual contracts.
• Top-Down Initiative: AI adoption is a top priority for CEOs and executives, making it easier to sell AI solutions.
• Sales Meetings: Deciders are typically at the CEO level or the most senior people in the firm.
• Importance of Buy-In: Requires buy-in from both top executives and the people implementing the tool.

9. Global Presence and Trust-Building

• Global Team: Model ML has teams in Hong Kong, Singapore, India, London, and New York.
• Building Trust: Trust is crucial for closing deals, especially since decision-makers could face consequences for wrong decisions.
• Strategies: Face-to-face meetings, detailed demos with real use cases, and customer-specific investments.

10. Lessons from Previous Companies (Fancy and Fat Llama)

• Enjoyment: Founders must enjoy building companies.
• Preparedness: Be prepared for a roller coaster experience with many ups and downs.
• Calmness: Experience from previous ventures helps in staying calmer during challenges.
• Perseverance: "If something logically makes sense, you should probably continue doing that thing, right? And not let anything stop you."
• Hiring: Focus on cultural fit and enjoyment of the work, not just CV credentials.
• Work Ethic: Initial period requires intense work ethic (7 days a week).
• Passion: Look for candidates who are passionate about building things.

11. Key Quotes

• Chaz: "In the last seven days, we've signed the same number of contracts as we signed in the whole of Q4."
• Chaz: "There is clear tangible value being driven by these products, and it's only going to get better and quickly."
• Arie: "If something logically makes sense, you should probably continue doing that thing, right? And not let anything stop you."

12. Technical Terms and Concepts

• Agentic System: An AI system that can act autonomously to achieve specific goals.
• Cognitive Architecture: A framework that mimics the cognitive processes of a human, including access to information and decision-making.
• OCR (Optical Character Recognition): Technology that converts images of text into machine-readable text.
• Function Calling: The ability of a language model to call external functions or APIs to perform specific tasks.

13. Synthesis/Conclusion

Model ML is experiencing rapid growth due to the increasing adoption of AI in the financial services industry. Their AI workspace, built on agentic systems and a cognitive architecture, automates key tasks, improves accuracy, and saves significant time for financial professionals. The founders emphasize the importance of perseverance, building trust with clients, and hiring individuals who are passionate about their work. The advancements in AI models, particularly in vision and function calling, are driving further improvements in their product and creating a significant competitive advantage.