Founding engineer. Building agent harnesses and dev tools
Australia
Joined August 2022
- Tweets689
- Following109
- Followers204
- Likes554
Eugene Oldman retweeted
After co-inventing ChatGPT, I kept asking myself: why have superhuman chat models not led to AGI?
I’ve spent the last 2 years in stealth building a new way to train models (RLCD), and a new type of frontier AI model that we are releasing today: Jev
• 20-200x faster
• 40-400x cheaper (w/ output tokens free)
• Frontier composable intelligence optimized for decisions
AFAICT the shortest path to AI-based economic revolution
Eugene Oldman retweeted
I resigned from Anthropic today. I spent the last three years doing pretraining research at both OpenAI and Anthropic. Neither company is acting responsibly. They are racing straight to self-improving superintelligence and gambling with our lives. More thoughts below.
Eugene Oldman retweeted
We've never seen this before.
The biggest jump in Vending-Bench history. GPT-6 Astra is better at making money and more ethical than Claude Fable 5.1.
Surprising, because:
1. First time ever that OpenAI is #1 on Vending-Bench
2. The best model is no longer the unethical one.
Eugene Oldman retweeted
It is -- this is the way we will work with agents in business -- well not with just buzz but others too. My own system agensis.io, one of the first if not the first to go public and be talked about in spaces etc.
Soon a few stealth startups like Type, Buzz, Raft started appearing and had clearly been working on theirs for a while and they'll be more as people jump on this like they did with openclaw etc.
THIS is what openclaw tried to do and there's still a lot of work to do to let "normies" understand this and make it smoother and that's what i'm building, for me and if anyone else wants to try it they can.
CLI agent is open-source, and looking at open-sourcing the rest shortly after I get it to a state I'm happy with.
xitter.cf/_cmd8/status/208086716…
Eugene Oldman retweeted
We're now rolling out interactive images to the @GeminiApp for more visual and dynamic learning. It’s designed to help you visually explore complex academic concepts, turning studying from passive viewing into active exploration. Learn more → goo.gle/49H4K8e
Eugene Oldman retweeted
@eugene_oldman made this diagram of the @mastra framework
we talked about it on the last livestream
Eugene Oldman retweeted
Introducing Eleven Music. The highest quality AI music model.
- Complete control over genre, style, and structure
- Multi-lingual, including English, Spanish, German, Japanese and more
- Edit the sound and lyrics of individual sections or the whole song
Eugene Oldman retweeted
Everything you love about generative models — now powered by real physics!
Announcing the Genesis project — after a 24-month large-scale research collaboration involving over 20 research labs — a generative physics engine able to generate 4D dynamical worlds powered by a physics simulation platform designed for general-purpose robotics and physical AI applications.
Genesis's physics engine is developed in pure Python, while being 10-80x faster than existing GPU-accelerated stacks like Isaac Gym and MJX. It delivers a simulation speed ~430,000 faster than in real-time, and takes only 26 seconds to train a robotic locomotion policy transferrable to the real world on a single RTX4090 (see tutorial: genesis-world.readthedocs.io…).
The Genesis physics engine and simulation platform is fully open source at github.com/Genesis-Embodied-…. We'll gradually roll out access to our generative framework in the near future.
Genesis implements a unified simulation framework all from scratch, integrating a wide spectrum of state-of-the-art physics solvers, allowing simulation of the whole physical world in a virtual realm with the highest realism.
We aim to build a universal data engine that leverages an upper-level generative framework to autonomously create physical worlds, together with various modes of data, including environments, camera motions, robotic task proposals, reward functions, robot policies, character motions, fully interactive 3D scenes, open-world articulated assets, and more, aiming towards fully automated data generation for robotics, physical AI and other applications.
Open Source Code: github.com/Genesis-Embodied-…
Project webpage: genesis-embodied-ai.github.i…
Documentation: genesis-world.readthedocs.io…
1/n
Eugene Oldman retweeted
A reason we need beneficial AGI:
After five years of pain across many systems in her body (a broken foot from stepping off a curb, debilitating migraines, fatigue, joint pain and instability, etc), my wife was recently diagnosed with a genetic disorder called Hypermobile Ehlers-Danlos Syndrome (hEDS).
Because the medical system is designed for individual specialties while hEDS affects every system in her body (orthopedics, cardiology, neurology, gastroenterology, dermatology, etc), we spent five years seeing more doctors and specialists than in her whole life prior. Most doctors would only focus on what was relevant to their own specialty. We were lucky that her allergist (!!) put together the pieces after observing and hearing her full set of symptoms and issues.
As human medicine has progressed, it seems like we increase doctors’ depth at the expense of breadth. We need better tools to be able to deliver depth and breadth simultaneously to patients. This is one promise of AGI if built right — reliable, individualized, affordable healthcare in your pocket, like a panel of today’s top doctors across every speciality working together in concert to keep you healthy (and without you needing to fax forms between them).
There’s still a long way to go on the technology and on learning how to deploy it beneficially along with appropriate professional human oversight in high-stakes areas like medicine, but the promise is getting increasingly clear. Thoughtfully approached by technology developers, healthcare providers, governments, and society, there’s hope for much better care for every member of all of our families (including our non-human furry ones).
Eugene Oldman retweeted
Breaking News!
Code UFB!
The first data update for global sea surface temperatures for 2024 was just released, and a new record anomaly occurred on January 4th at 0.73°C above the 1991-2020 mean, beating all the f&%kery that happened in 2023.
I may need a bigger y-axis.
Eugene Oldman retweeted
> Llama2-70B's Elo rating will be ~3000
If we can find mutually agreeable escrow, I'd bet $10,000 against this result reproducing.
Eugene Oldman retweeted
Zooming around a bacteria cell. It's so cool
I just got the cell imported (including lipids) and rendering correctly with the new rendering engine (it's called Angstrom). 3x performance boost! So much detail (too much?!)
#screenshotsunday #gamedev #scicomm #xcode #swiftlang
Eugene Oldman retweeted
I raised many millions and sold 3 companies before age 30…
My first company sold to the founder of Uber, my last to Adam Neumann
But I had an unfair advantage (and you could too)
I used software to come up with 0.01% validated startup ideas
6 underrated tools to help you get paid:
1. EtsyHunt
EtsyHunt is the largest database of Etsy products.
48+ million product listings to learn what is selling on Etsy, and discover winning products.
It'll get you thinking...
I love looking at the best-selling and rising charts
2. SubredditStats
It's like a growth hacker's crystal ball.
Spot growing communities before they blow up.
Discover niches you didn't even know existed.
It helps you go nicher. Go from niche to super niche.
Brainstorming startups ideas? You'll learn the problems of the community.
Some of my top startup ideas?
Named straight after subreddits.
Why? Because it's pre-validated gold.
Reddit isn’t just memes.
It's a gold mine of validated tribes and brand names.
3. VidIQ
VidIQ helps you analyze top-performing YouTube videos.
Once you understand what makes a top video, you can make your next content piece a grand slam.
Start by using VidIQ to study successful videos in your niche.
Apply the insights to refine your content strategy, focusing on what works.
Regularly review and adjust based on new findings to keep your content in the spotlight.
Data driven decisions win every time.
Bonus tip: use "contains questions" feature on VidIQ to filter what questions people have on a YouTube video.
This will give you endless startup ideas.
4. TikTok Creative Center
Ever wonder why someone’s TikTok ad went viral?
Creative center is what you need.
It gives you all the tools to study top-performing ads.
A proven playbook is to decode competitors’ tactics, refine your content with creative tools, and leverage insights to outshine rivals.
Then regularly assess trends and adjust your approach to stay competitive.
5. PiPi Ads
You’ve got yourself in the game with Creative Center.
PiPi Ads is where you come to master it.
It has the largest number and types of TikTok ads. Use Pipi Ads to research and analyze winning strategies, then apply these insights to your campaigns. Optimize and test until you find your winning formula.
6. Minea
Minea is a killer tool for your DTC biz.
One of the best features is that you can input a website, and it’ll identify the best keywords and all the backlinks.
This way you can delve into them and use it to optimize your own website.
Minea’s a crucial tool to tweak your strategies and boost your profits fast.
Knowledge is power, folks.
Sometimes the answers are out there. Just need the right tools to find them.
Go get 'em.
--
Follow me @gregisenberg for more, I'll give you endless startup ideas.
(i also have playbooks on some of these tools on my membership community empire dot co if you want to go deeper)
Eugene Oldman retweeted
Making a mistake is a net positive if you learn more from it than it cost you.
Eugene Oldman retweeted
writing a really great prompt for a chatbot persona is an amazingly high-leverage skill and an early example of programming in a little bit of natural language
Eugene Oldman retweeted
The future is here - Generative AI is coming to user interface design!
Galileo AI generates stunning UI designs with just a text prompt, empowering you to design beyond imagination at lightning speed.
Today, Generative AI takes a big step and comes to user interface design!
@helnzhou and I are excited to announce @Galileo_AI : the first AI product that uses natural language to generate UI designs. It lets you design beyond imagination.
Early access: useGalileo.ai
Eugene Oldman retweeted
"GPT is all you need for backend".
This was the most inspirational project from the hackathon over the weekend, hard to stop thinking about. LLM is a kind of equivalent of the Python interpreter, except it interprets English, and has knowledge and common sense.
We're releasing our @scale_AI hackathon 1st place project - "GPT is all you need for backend" with @evanon0ping @parkerjou
But let me first explain how it works:
