How Union Square Ventures Built an AI Brain for Venture Capital

TL;DR: Today we’re releasing a new episode of our podcast AI & I. I go in depth with Matt Cynamon, who runs experiments with AI at Union Square Ventures. We get into how AI enables generalists to build and conduct live demos of the suite of AI tools he’s building for USV. Watch on X or YouTube, or listen on Spotify or Apple Podcasts.

Union Square Ventures is building an AI operating system to support their investment team. 

But it’s not what you think: It’s a constellation of AI tools that captures and synthesizes the firm's collective wisdom. It’s evolving every day, and Matt Cynamon is the mad scientist in charge.

Matt calls himself a “regular” at USV. In practice that means he’s responsible for running experiments with AI for the firm. As an inherently curious person with the professional obligation to tinker, he’s built a suite of tools for the firm, including: 

• The Librarian, a chatbot trained on around 15,000 articles from USV’s blog
• Portfolio Tracker, a GPT that analyzes the investments made by the firm
• Meeting Notes, a tool that makes it possible for team members to interact with meetings  

I sat down with Matt to talk about how AI is enabling him to bring his ideas to life as a generalist, get demos of the tools listed above, and exchange notes on all the other projects he has in the works at USV. We try out an art project at USV’s office called The Dream Machine, which generates art from conversations. And we edit actionable insights extracted by an AI from meetings at USV and prepare them to be posted on the firm’s X handle live on the show. Here’s a link to the episode transcript.   

This is a must-watch for anyone interested in riding the AI wave by learning how to ship useful products quickly.

Watch on X or YouTube, or listen on Spotify or Apple Podcasts. 

If you want a quick summary, here’s a taste for paying subscribers:

Sponsored by: Every

Tools for a new generation of builders

When you write a lot about AI like we do, it’s hard not to see opportunities. We build tools for our team to become faster and better. When they work well, we bring them to our readers, too. We have a hunch: If you like reading Every, you’ll like what we’ve made.

• Automate repeat writing with Spiral.
• Organize your computer with Sparkle.
• Write something new—and great—with Lex.
• Deliver yourself from email with Cora.

Want to sponsor Every? Click here.

How AI empowered Matt to learn by doing 

Matt inherited his affinity to tinker from his father, “the type of person to always had a broken-down piece of machinery, like a car or a computer, that he was building from scratch.” However, until recently, his broad but shallow knowledge base—the burden of a “generalist”—held him back from executing his ideas. This is how AI has leveled the playing field: 

• Generalists become creators. AI enables people without specialized skills to bring projects to life, like Matt—who isn’t a developer—who used LLMs to learn how to connect the APIs between a custom GPT and a live spreadsheet. He points to “develop[ing] the skill alongside seeing your creativity come to life” and “the feedback loops [being] so fast” as key benefits of AI.
• Understand as you create with AI. AI tools have created a new way of learning by allowing people to experiment even before they fully understand underlying concepts. According to Matt, you can “build your skills” by using AI tools to execute your vision, and subsequently “take it apart to see what it is that you just did.”

The suite of AI tools USV uses for better investing 

Matt’s initial idea when he started working full-time on AI projects at USV was to build an “all-in-one monolithic super app,” but he soon realized that the best way to create an app that’s helpful to the firm today would be to “use tools that already exist, break them down into their component parts, and build individual agents that do different things.” These are a few of the tools that Matt has built at USV, along with demos live on the show:

The Librarian 

The Librarian is a “chatbot built on top of [USV’s] entire history of writing,” a particularly useful tool given the penchant the firm’s partners have for writing. In the process of building it, Matt “realized how powerful metaphor can be in explaining what it is that these chatbots can do for us.” He originally named it “Conversations,” because he wanted the partners to be able to question the chatbot about USV’s previous articles as they wrote new blog posts, and when that didn’t quite work, “At The Edge,” because the bot could help “synthesize and advance [their] thinking.” But it wasn’t until he landed on “The Librarian” that the chatbot caught on internally.  

Matt initially built the Librarian as a standalone application, with help from No Code MBA and Ben’s Bites, but since maintaining a custom user interface is troublesome, he’s decided to “break down all the component parts of The Librarian to the individual things that [it] was doing” and make them each a separate GPT that handled a specific function that the original Librarian performed.

One of the GPT’s functions that Matt demos live on the show involves prompting The Librarian with the name of a company USV is interested in investing in, and instructing the LLM to pull out relevant USV blog posts that might pertain to the company. He says this helps USV prove to entrepreneurs that: “Hey, we're not just chasing the hot thing, you are actually what we've been looking for for a long time, and we have the receipts.” This function, which typically takes partners a lot of time, can now be completed efficiently using the chatbot.

Portfolio Tracker

The next tool Matt demos is a spin-off of The Librarian. Portfolio Tracker is a chatbot that “categorizes all the companies [USV’s] invested in and when [they invested],” including information around the firm’s “ownership percentage” in them. Matt screenshares a chat where the GPT created a chart of all the companies USV has invested in education, search, and AI, dating back to 2005, which provides valuable context to a partner looking to invest in a company that operates in any of those areas.

Meeting Notes

Meeting Notes is a chatbot trained on “topical meetings” USV conducts around subjects like “AI, crypto, climate, and a miscellaneous one where [they] hash out ideas and discuss specific companies.” Internally, the tool enables USV team members to “ask questions to the meetings.” Matt explains that Meeting Notes wins over other specialized note-taking tools—like Granola and NotebookLM—because of the ease with which it can be accessed. “If any member of USV opens up ChatGPT or Claude on their phone, there's a conversational interface with every part of the business in the left-hand column that they have access to right there, without having to toggle between a million different apps,” he says. 

This is also the chatbot that generates snippets of these meetings—named “Overheard at USV”—for the firm to post on X. The idea for this was born from a concern that USV was not creating enough content, so they used the “great stories happening in meetings” as raw material, “capturing” and “curating” them for the X posts. Matt notes that it also serves as an effective way to “let people into [USV]’s processes earlier.” 

The Dream Machine

The last AI project that Matt demos is a tool that can listen to a conversation, generate moving images that represent the thoughts of the people involved, and project them onto a large television screen. Matt explains how the Dream Machine works on the back end: “[O]ne setting [of the television] is hardwired directly into a piece of machinery that’s locally running a visual diffusion model, that’s connected to the Whisper API, that listens to what we are saying, and then in real time generates moving images based on what we're talking about.”

The Dream Machine is developed by a member of a Bright Moments DAO, in which USV is an investor. “This is one of the first truly native AI art forms that couldn’t exist in a world prior to AI,” Matt says. 

Matt wrapped up the demos by explaining that USV has kept their AI tools in house so far, noting there's a “certain level of tolerance for information being wrong” internally. He added that the team is working on improving these tools’ reliability to make them suitable for “public consumption.”

AI tools for the future of venture capital

Here are a few projects that Matt has in the pipeline at USV:

• A publicly available custom GPT trained on everything USV has written. “[I]t’s highly likely that if you come to our website, at some time before the end of the year, all that's going to exist is The Librarian.”
• A tool that surfaces relevant USV blog posts when founders share their pitch decks. “[T]hat enables you to come to the meeting ready…kind of get[ting] aligned before you even walk in the door.”
• A tool called “Playbooks,” that automatically transcribes and makes notes of meetings where “tricky situations” portfolio companies go through are discussed—like if a CEO wants to step down—with the objective of building a knowledge base of problem-solving approaches. 

You can check out the episode on X, Spotify, Apple Podcasts, or YouTube. Links and timestamps are below:

• Watch on X
• Watch on YouTube
• Listen on Spotify (make sure to follow to help us rank!)
• Listen on Apple Podcasts

Timestamps:

1. Introduction: 00:00:52
2. How Matt became in charge of everything AI at USV: 00:01:56
3. How AI empowers generalists to be creators: 00:06:22
4. The Librarian, a chatbot trained on everything USV has published: 00:10:41
5. Portfolio Tracker, an AI tool to track USV’s investments: 00:21:09
6. The AI projects that Matt has in the pipeline at USV: 00:27:21
7. Meeting Notes, USV’s AI note-taking tool: 00:34:33
8. Prompting AI to generate a post for USV’s X handle: 00:44:57 
9. Why it’s important to diversify ownership over data: 01:00:20
10. The Dream Machine, AI that generates images from conversations: 01:03:20

What do you use AI for? Have you found any interesting or surprising use cases? We want to hear from you—and we might even interview you. Reply here to talk to me!

Miss an episode? Catch up on my recent conversations with star podcaster Dwarkesh Patel, LinkedIn cofounder Reid Hoffman, a16z Podcast host Steph Smith, economist Tyler Cowen, writer and entrepreneur David Perell, founder and newsletter operator Ben Tossell, and others, and learn how they use AI to think, create, and relate.

If you’re enjoying my work, here are a few things I recommend:

• Subscribe to Every
• Follow me on X
• Subscribe to Every’s YouTube channel

Dan Shipper is the cofounder and CEO of Every, where he writes the Chain of Thought column and hosts the podcast AI & I. You can follow him on X at @danshipper and on LinkedIn, and Every on X at @every and on LinkedIn.

We also build AI tools for readers like you. Automate repeat writing with Spiral. Organize files automatically with Sparkle. Write something great with Lex.