How a Network of Agents Can Run a Hedge Fund

What happens to the analyst’s job when one AI agent can run an entire team of research agents?

In this episode of Invest With AI, Brett Caughran sits down with Robert Yang, co-founder and CEO of Fundamental Research Labs (FRL), to explore where AI-driven investment research is headed, and why Brett is joining FRL to help build it.

Robert started as a computational neuroscience professor at MIT before leaving academia to build AI agents. His path ran through a 1,000-agent society in Minecraft and ended at Shortcut, an Excel agent now used by buy-side analysts.

FRL is building beyond the spreadsheet.

The goal is a team of agents that works like a hedge fund, with analysts researching companies and a portfolio manager coordinating the work, built so investors can adapt it to their own process.

But as agents take on more of the analyst’s work, a bigger question emerges: Can AI hold a real investment view, or will it always agree with whoever prompted it last?

We get into:

  • From MIT to Minecraft to Wall Street

    • Robert explains why he left MIT to build agents, and what FRL learned when 1,000 AI agents in Minecraft organized a search party nobody programmed.

  • Why Hedge Fund Analysts Became the Customer

    • Shortcut went viral before the models were ready. A rigorous internal benchmark helped close the gap, and analysts with no tolerance for errors became the users who felt the difference most.

  • From 50 Agent Threads to One Agent That Runs the Team

    • After managing 50 agents across seven monitors, Robert sees the future as talking to one agent that coordinates the rest.

  • Teaching an Agent to Hold an Investment View

    • Today’s models are built to agree with you. Robert outlines a path through custom skills, post-training and continual learning toward agents with grounded conviction.

  • When Analysts Start Working More Like PMs

    • As digital analysts arrive, roles may move up a level. Brett shares how his work at FRL continues what he’s done for years: training analysts, now including digital ones.

Chapters (Timestamps)

[00:00] Intro
[00:43] Why Robert Wanted to Build AI Before It Was Popular
[08:40] Running an MIT Lab and Questioning the Work
[14:05] The Agent Research That Pushed Him to Leave
[19:20] Project Sid: Building a Society of 1,000 Agents
[22:18] The Search Party Nobody Programmed
[26:58] From Minecraft to Excel Agents for Wall Street
[33:05] Going Viral Wasn’t Enough: Making Shortcut Work
[37:28] Why Hedge Fund Analysts Became the Customer
[41:51] From an Excel Agent to an Investment Team
[45:51] Why Nobody Wants to Manage 50 Agent Threads
[50:12] How Close Are We to a Digital Hedge Fund Analyst?
[53:52] The Problem with AI That Always Agrees
[56:50] Teaching an Agent to Hold an Investment View
[1:04:29] When Analysts Start Working More Like PMs
[1:06:29] Brett’s Role: Training the Digital Analyst
[1:09:07] If AI Frees Up Your Time, Where Does It Go?
[1:11:34] What This Means for FE and Invest with AI

Follow Invest with AI on:

Spotify: https://open.spotify.com/show/033xcEEovVViS7hIYwNuGZ

Apple Podcasts: https://podcasts.apple.com/us/podcast/invest-with-ai/id1896918892

Next
Next

Can AI Replace Traditional Expert Networks? The Future of Primary Research with Qualitate