Harshil Agrawal’s Post

Stack Overflow signed a deal with OpenAI, announced Monday. Two things to think about: 1. Contributors aren't getting compensated for the value of their contributions. I'm not making a moral or legal point, just a practical one. It kinda sucks creating other people's value for free when they flaunt it. At least with open source generally, you're sorta insulated from the monetization by anonymity of reuse. At the same time, any community that doesn't ban you is always going to be a cool hang if you're a halfway interesting contributor or person. 2. People worry about LLM generated content feeding the LLMs and making the LLMs regress. Don't worry about that. It won't happen. There is a different problem. LLM makers won't let their LLMs regress on metrics that matter (no matter how bullshitty the metrics). They will have to spend more time and money filtering their training data sets as they become less human and more generated. That actually sounds like an activity OpenAI can saddle SO with. "We'll pay you for ongoing use of your data so long as your data improves." The fun part is that it will take time and GPU money in "pre-training" to prove that iterations are improving. In short, LLMs won't regress. They will get more expensive. Qualified training data will cost more. This is a critical business insight in this space, and is probably very different than whatever narrative you've bought into. If you need this kind of insight across your business, I am available to help you, from consultant and projects to full time. Message me. H/T Axel C., who is really good at LinkedIn and will probably end up monetized without compensation if he isn't already. #WrittenByMe

Swarm from OpenAI: This is an interesting preview of AI agents having the capability to work together across networks, something I believe is coming and will unlock all sorts of interesting use cases and new value. https://lnkd.in/eUcFMY8v

From OpenAI versus Stack Overflow To OpenAI and Stack Overflow The technological advancements happen when you replace versus with and. Read more at: https://lnkd.in/giPsnzdR

🌊 LaVague now integrates OpenAI #GPT4o model! Timing coincides perfectly with our latest version of LaVague, that now has a World Model built-in to perform actions on the web using high level objectives like “Ask Llama 8B ‘What is love?’”. We have provided integration of OpenAI GPT-4o model into LaVague, our Large Action Model framework for AI Web Agents. 🤯 First results: it’s super fast (compared to all previous vision models we tried before) while still being good enough to perform web actions. So if you want to play with trying/building your AI Web Agent using Large Action Models, you can look at our: GitHub: https://lnkd.in/e22QkuQz Google Colab: https://lnkd.in/eCE7XnVZ Quicktour: https://lnkd.in/eacqKHuN

Open-source LLMs are in pace to the commercial LLMs in transforming the digital landscape with AI, and there are multiple ways to serve these models. One of the most efficient and flexible approaches is serving them via an API (e.g., using the FastAPI framework), by making the API compatible with the OpenAI client, it can easily integrate with a variety of orchestration frameworks, leveraging the popularity of OpenAI's platform. In this recent article, you'll get a quick-start on: 1️⃣ Building an API to serve open-source LLMs that is fully compatible with the OpenAI client. 2️⃣ Deploying LLMs via the vLLM framework for faster and more efficient performance. 3️⃣ Implementing basic rate limiting to control the flow of API requests and maintain system stability. 4️⃣ Setting up API key authentication to securely access the API. 5️⃣ Managing custom API keys through a simple Streamlit application. 6️⃣ Enabling user authentication in the Streamlit app using Azure email communication. 🔗 Check out the full article below! https://lnkd.in/dUr2jsrD

Engineers aren't very good at writing prompts. There're no concrete requirements, or easily definable success or failure cases (unless the output is REAL garbage). It's also time-intensive. You have to sit down and think, and refine your prompts, and iterate continuously based on experimentation and user feedback. This is something that requires _way_ more patience than most engineers have. QA, Managers, PO's, are all better at this than engineers are. When I was working in Ai as the generative boom was starting, there was a constant sense of tension over the question of "who owns the prompt?". I wanted to create a system that freed the prompts from the engineers, but in a way that was safe for the product (i.e. no-one could accidentally delete a row in a table and then crash the app), and allowed for fluent versioning and independence from any single Ai provider. This is where Easybeam's Portals came from, they allow you to, - Change your Ai provider, prompts, and Ai config without any code - Have role-based access to your prompts / Ai config - Observe logs to ensure your Ai's not up to any funny business - Generate analytics to see trends in performance and quality It's a start, but I hope this can save teams a serious headache or two in the future, and let them leap-frog off of my work and the features they're building

Lately, I've been experimenting a lot with LLMs, and I noticed an interesting trend in the adoption of OpenAI API compatibility. Google is the latest to add OpenAI API compatibility for its Gemini LLMs: https://lnkd.in/efRfND5G ❓ Wondering why? Why not just stick with their own solution? Fast and friction-free onboarding is becoming the key to adoption in the fast-paced race of large language models (LLMs). OpenAI’s API has quickly become the de facto standard for integrating LLMs, thanks to its simplicity, early market presence, and broad ecosystem. Almost anyone who’s experimented with LLMs has likely started with OpenAI’s API, which provides a seamless experience from setup to getting the first response back, setting a high bar for others in the industry. For any company looking to compete in the LLM space, making the switch to their integration as easy as possible is critical. This is why we’re seeing so many providers embracing OpenAI API compatibility: xAI, Perplexity, and others already offer compatibility, and Hugging Face can serve any model with an OpenAI-compatible endpoint. ❓ Why is this important? This compatibility lets developers use new models simply by changing the endpoint in their OpenAI client—no additional code changes are required. Even better, it means developers can leverage existing tools, plugins, and the broader OpenAI ecosystem they’re already comfortable with. This friction-free integration removes barriers to trying out new solutions, making it easier than ever for customers to explore new options without leaving behind the tools they rely on. 💡 How you can translate this to your business? A smooth customer experience and easy onboarding can be a game-changer for any business. Here are a few ideas that you can take from this observation: 1️⃣ Lower barriers to entry Make it as easy as possible for customers to try your product. If they can integrate or onboard quickly, they’re more likely to stick around. 2️⃣ Adopt familiar standards: Wherever possible, align your product with industry standards that your customers are already familiar with. This reduces the learning curve and builds instant trust. 3️⃣ Focus on flexibility and compatibility: Design your product to work seamlessly with other tools your customers might already be using. Flexibility empowers users to experiment with your solution without fear of reworking everything. A great user experience isn’t just a feature—it’s a strategy. When customers can test your product with minimal effort, you create a powerful first impression and open the door for broader adoption. #developerexperience #dx #ux #openai #gemini #llm #observation

Some fact findings about AI: https://lnkd.in/gwcPNUPP A simple go/no-go test shows that, with our intellectual property (IP), a copyrighted Chinese-English multilingual metadata, we can do what artificial intelligence (AI) can't do in data analytics, NOW.

Some fact findings about AI: https://lnkd.in/gwcPNUPP A simple go/no-go test shows that, with our intellectual property (IP), a copyrighted Chinese-English multilingual metadata, we can do what artificial intelligence (AI) can't do in data analytics, NOW.