Unlocking AGI With Visual AI Agents | Joseph Nelson, Roboflow
The Peel with Turner Novak
Turner Novak
4.6 • 11 Ratings
🗓️ 19 December 2024
⏱️ 118 minutes
🧾️ Download transcript
Summary
Joseph Nelson is the Co-founder and CEO of Roboflow, making the world programmable by building computer vision tools for developers and enterprises.
We talk about how computer vision creates a new paradigm to program the world, and how visual AI is the missing piece of AGI.
Joseph also shares multiple live product examples, how computer vision unlocks new data sources, lessons from Stripe and Palantir, building business models in developer tools, his experience working with David Sacks, and developer marketing tactics and how Roboflow consistently gets to the front page of Hacker News.
Timestamps:
(00:00) Intro
(03:34) Computer vision is the missing piece for AGI
(05:59) Vision as a new paradigm to collect data
(10:55) Live examples of computer vision
(13:45) How a Magic Sudoku solver app led to Roboflow
(18:13) Using computer vision for automation
(24:49) Computer vision in sports
(27:02) How vision unlocks new data sources
(28:24) Inside developer tool business models
(33:32) The "Collison Install" and hands-on customer service
(36:45) When to adopt Palantir's Forward Deployed Engineers
(43:44) Why AI companies need to combine PLG and enterprise sales
(50:12) Advice on developer marketing
(52:30) Roboflow's greatest hits on Hacker News
(01:02:19) Benefits of David Sacks as AI & Crypto Czar
(01:05:32) Why all new technology has bad actors
(01:07:07) Why over-regulation holds back innovation
(01:12:01) How to get on the front page of Hacker News
(01:19:43) Multi modality, time recognition, and agentic vision
(01:28:36) Image-to-image prompting
(01:30:42) Growing up in Iowa
(01:32:20) Making TI-84 calculator games in high school
(01:36:32) Pioneer: hunger games for startups
(01:40:16) Why Roboflow does weekly Ship Lists + Ship and Tell
(01:42:46) Hiring former founders and "full stack people"
(01:45:16) Designing a bottoms-up organization while scaling
(01:50:35) Why candidates build with Roboflow in hiring process
(01:55:08) Hiring someone to help with the podcast
Referenced:
Robowflow: https://roboflow.com/
Roboflow Universe: https://universe.roboflow.com/
Paint.wtf: https://paint.wtf/
Roboflows NeurIPS Presentations: https://blog.roboflow.com/neurips-2023-papers-highlights/
Careers at Roboflow: https://roboflow.com/careers
Follow Joseph:
Twitter: https://x.com/josephofiowa/
LinkedIn: https://www.linkedin.com/in/josephofiowa
Follow Turner:
Twitter: https://twitter.com/TurnerNovak
LinkedIn: https://www.linkedin.com/in/turnernovak
Subscribe to my newsletter to get every episode + the transcript in your inbox every week: https://www.thespl.it/
Transcript
Click on a timestamp to play from that location
| 0:00.0 | To really have AI realize its full potential, it needs to understand the real world. |
| 0:05.0 | And understanding the real world means having a sense of visual understanding and reasoning and capability, |
| 0:11.0 | being able to understand context, navigate environment, make decisions, and as exciting and prominent as language is, vision is humans first sense. |
| 0:23.6 | Like we actually had our sense of sight before we even had the ability to construct language. |
| 0:28.7 | And so every product, every service will also have the ability to make visual understanding and visual decisions. |
| 0:35.3 | And that starts with computer vision and visual AI. |
| 0:38.5 | Welcome to the Peel. I'm your host, Turner Novak, founder of Banana Capital. Today's guest is |
| 0:43.6 | Joseph Nelson, co-founder and CEO of RoboFlo, building computer vision tools for developers and enterprises. |
| 0:50.1 | Giving a camera, the intelligence, and deploying it and relying on it. |
| 0:55.0 | Previously, you know, would maybe be like a quarter's worth of work and not so long ago, like a PhD thesis. |
| 1:01.0 | And now with current tools, you can do that in an afternoon. |
| 1:04.0 | RoboFle has over 1.3 million active developers, |
| 1:08.0 | hundreds of millions of pre-trained labeled open source images and |
| 1:11.2 | models. We talked through what's coming next with visual AI agents, and Joseph even pulls up |
| 1:16.3 | some live examples. How many whenever a package shows up on my doorstep? How many whenever there |
| 1:21.3 | looks like there's going to be a jam on my line? Customer service lessons from Stripe. |
| 1:25.1 | Patch would just show up at your office, get on your computer, and install Stripe. |
| 1:30.3 | Learnings from Palantir's forward-deployed engineers. |
| 1:33.3 | I think startups fail to mimic this because a lot of them don't realize. |
| 1:37.3 | Developer tool business models |
| 1:39.3 | Is our self-serve plan its own line of business or is it lead gen for our enterprise business? |
| 1:47.4 | His experience working with David Sachs. |
... |
Please login to see the full transcript.
Disclaimer: The podcast and artwork embedded on this page are from Turner Novak, and are the property of its owner and not affiliated with or endorsed by Tapesearch.
Generated transcripts are the property of Turner Novak and are distributed freely under the Fair Use doctrine. Transcripts generated by Tapesearch are not guaranteed to be accurate.
Copyright © Tapesearch 2026.