George Kanellopoulos

I built thinkllm.dev, a knowledge graph of 1,500+ AI models. Find the right model for your use case, explore capabilities, and compare models. Dedicated sections for 800+ research papers and 3,400+ glossary terms. Free, no signup required.

@DanielSmidstrup Sanity...

USA has ChatGPT USA has Grok USA has Claude USA has Gemini China has DeepSeek China has Qwen China has Kimi China has MiniMax Europe has?

We have so much to tell you, we needed another account. Follow @OpenAINewsroom for company news, updates, and the occasional clarification.

We’re partnering with Condé Nast to deepen the integration of quality journalism into ChatGPT and our SearchGPT prototype. openai.com/index/conde-na…

Breaking news—we have a news handle.

@QCompounding Watch The Sniffing Accountant and The Stock Tip from Seinfeld and you are good to go ☺️

7 Documentaries that will teach you more about money: 1) Inside Job

Have you heard of the 12-Factor App? The "12 Factor App" offers a set of best practices for building modern software applications. Following these 12 principles can help developers and teams in building reliable, scalable, and manageable applications. Here's a brief overview of each principle: 1. Codebase: Have one place to keep all your code, and manage it using version control like Git. 2. Dependencies: List all the things your app needs to work properly, and make sure they're easy to install. 3. Config: Keep important settings like database credentials separate from your code, so you can change them without rewriting code. 4. Backing Services: Use other services (like databases or payment processors) as separate components that your app connects to. 5. Build, Release, Run: Make a clear distinction between preparing your app, releasing it, and running it in production. 6. Processes: Design your app so that each part doesn't rely on a specific computer or memory. It's like making LEGO blocks that fit together. 7. Port Binding: Let your app be accessible through a network port, and make sure it doesn't store critical information on a single computer. 8. Concurrency: Make your app able to handle more work by adding more copies of the same thing, like hiring more workers for a busy restaurant. 9. Disposability: Your app should start quickly and shut down gracefully, like turning off a light switch instead of yanking out the power cord. 10. Dev/Prod Parity: Ensure that what you use for developing your app is very similar to what you use in production, to avoid surprises. 11. Logs: Keep a record of what happens in your app so you can understand and fix issues, like a diary for your software. 12. Admin Processes: Run special tasks separately from your app, like doing maintenance work in a workshop instead of on the factory floor. Over to you: Where do you think these principles can have the most impact in improving software development practices? -- Subscribe to our weekly newsletter to get a Free System Design PDF (158 pages): bit.ly/3KCnWXq

@matt_gray_ The Sharia law makes Dubai a forbidden destination for building a life, if of course ethics and human rights is part of one's worldview.

My goal in life: visit all 195 countries in the world before I die. But after going to 100+ cities in the last 8 years, 1 place has BLOWN all the others out of the water. Here’s why this city is perfect for high-value founders who want to build their empires:

Wouldn't it be great if you could dynamically change the conversation name in the ChatGPT UI as part of building custom GPTs? #OpenAI #ChatGPT #CustomeGPTs @OpenAI

Learn the fundamentals of Probability and Statistics from Brown University. They provide amazing visualizations to help you learn. And it’s 100% FREE 🔗seeing-theory.brown.edu

🚨 Massive AI Security Release 🚨 @NIST just put out the best AI Security Publication that I've ever seen. It is 106 pages of deep, technical content. It references real-world practical attacks. In this thread is the link and I'm going to cover a few highlights. 👇

I just launched CloudCatalog! An open source tool to help you document your AWS architectures. 🔎 Discover resources 📄 Document resources 👥 Assign owners ⚡️ Powered by markdown 🤩 And more... If you find it useful, a retweet would be awesome! ♥️🌎 cloudcatalog.dev

@tunguz Call Harold...

Looking for an intern [unpaid] for a side project. Here are some of the must-have prerequisites: - 5 years of experience with CUDA 12 - 7 years of experience working with DGX GH200 - 10 years of experience with Vision Pro - 12 years of experience with GPT4 Ping me if interested

Link to OpenAI’s full Prompt Engineering Guide: platform.openai.com/docs/guides/pr…

🦜Retrieval-Augmented Generation: From Theory to LangChain Implementation 👏We've really been enjoying @helloiamleonie's articles lately Check out this fantastic blog covering the basics of RAG, the theory behind it, and how to use it in practice Blog: towardsdatascience.com/retrieval-augm…

As a researcher, I have two kinds of ideas: the bad ones, and those that other people had 5y ago.

The Hitchhiker’s Guide From Chain-of-Thought Reasoning to Language Agents Very nice summary of CoT reasoning, foundational mechanics underpinning CoT techniques, and their application to language agent frameworks. There is an interesting discussion about generalization, efficiency, customization, scaling, and safety related to CoT, language agents, and LLMs in general. CoT is probably one of the most effective prompting techniques available to LLM developers today. It's worth learning everything you can about CoT and how to best leverage it. paper: arxiv.org/abs/2311.11797 repo: github.com/Zoeyyao27/CoT-…

What happened at OpenAI? > Nov 2 -> Sam was in the room, when the team demonstrated the next big improvement. 3 times before in OpenAIs history, most recently with GPT-4, they’d pushed back the veil of ignorance and pushed forward the frontier of discovery. As he watched the latest advance, he was already planning for the next moves: the funds that would have to be raised, the resources that would have to planned for > Nov 4 - Ilya was unsettled. They’d reached a threshold of autonomy that was concerning, while the alignment team was still just adding capability instead of emotion, actual love for humanity. They needed more time to figure out the research pathway instead of hurrying to deploy product. > Nov 6 - OpenAI dev day goes well. Lots of kudos. They release the announced products over the next several days After dev day, Greg and Sam are in full on fund raising mode. They’re targeting 90 billion valuation, a 3x lift from 30 billion. OpenAI recruiters are already calling Google employees with 10-20 mil 4 year packages telling them if they join now they’ll make it in before the valuation increase. Nov 9 > Greg meets Emanuel Macron for a photo op, afterwards take investor meetings in Paris. OpenAI is now at the sovereign wealth fund, European family money sized investment Nov 13 > in an interview with the Financial Times in London, Sam confirms for the first time that work has begun on GPT5 and that he’s fundraising. Nov 14 > OpenAI pauses ChatGPT subscriptions, as they’re totally out of capacity for inference. Usage has spiked post dev day. Nov 16 > APEC CEO summit, Sam confirms the Nov 4 next step up discovery By this point Ilya is pissed and scared. The fundraising tour has caught him by surprise. The usage numbers have spiked and now they don’t have enough servers for the research team, meaning they have to fundraise again to support research. Events are starting to move beyond him. Nov 17 10 am > board meeting starts. Ilya produces 2 23 year old staff engineers (rumor) to describe progress and dangers to the board. Turns into a battle when he says they shouldn’t be fundraising and shouldn’t be expanding and that Sam was out on tour without board agreement. Sam points out that as the CEO he has to stay ahead of the team and manage news flow in advance of the team’s needs. “We’re going to need more money for more data centers” Ilya has a fit. Things come to a head. A vote is taken. Sam is out. The board demotes Greg, who is not at the meeting as Sam is authorized to cast his vote.

We're rolling out new features and improvements that developers have been asking for: 1. Our new model GPT-4 Turbo supports 128K context and has fresher knowledge than GPT-4. Its input and output tokens are respectively 3× and 2× less expensive than GPT-4. It’s available now to all developers in preview. 2. Assistants API and new tools (Retrieval, Code Interpreter) will help developers build world-class AI assistants within their own apps. 3. The platform is becoming multimodal. GPT-4 Turbo with Vision, DALL·E 3, and text-to-speech are all now available to developers. Oh… and we’re doubling GPT-4 rate limits. openai.com/blog/new-model…

If you 1) already know RAG basics, and 2) want to become a superstar AI engineer, then learn to build advanced RAG from scratch 💪 Excited to launch a new category of @llama_index tutorials on this exact topic 🧑‍🏫🪐: ✅ learn advanced IR/AI concepts ✅ tackle more complex user queries ✅ Improve accuracy + reduce hallucination In this first part, we teach you how to build a RAG-Fusion pipeline: enables dynamic retrieval over complex questions. 1️⃣ Query generation/rewriting 2️⃣ Ensemble retrieval 3️⃣ Reciprocal Rank Fusion All these steps are based off core retrieval principles, designed to improve precision/recall for different user queries. By learning this, you'll gain the tools/intuitions for optimizing your RAG systems 💡 This is based off our existing `QueryFusionRetriever` which is in turn based on the awesome RAG-Fusion repo by (@Raudaschl). ❓Who is this tutorial for ❓ If you want something time-constrained/works out of the box, this may not be for you (check out our Quickstart instead!) On the other hand, if you 1) have deeply custom requirements, 2) are running into issues with naive RAG, 3) find out about our 100+ guides and have no idea why things work, then I’d highly recommend this to strengthen your understanding of retrieval! docs.llamaindex.ai/en/latest/exam…