All about AI

Paul wonders if the increasing use of LLMs and agents in software development will lead to a reduction in the need for new frameworks and extra developer-focused abstractions.

Dependency-free, full-text and vector search engine for all JS runtimes, with typo tolerance, filters, facets, stemming, and more.

In the past, programming and machine learning were two distinct domains with unique languages and tools. As machine learning is so much more math-heavy and technical than programming languages, it needed a specialized skill set for implementation.

Today, we are excited to introduce the first two models of the next generation of Llama, Meta Llama 3, which are now available for widespread use. This release includes pre-trained and instructionally fine-tuned language models with 8 billion and 70 billion parameters, capable of supporting a wide range of use cases.

In this session, Todd Anglin shows you how GitHub Copilot works, some best practices, and what AI can realistically do for developers in the real world, including unique capabilities available in Visual Studio and Visual Studio Code. AI-assisted developer tools, like GitHub Copilot, will change how you write code; but only if you are curious, creative, and you continue to think critically.

Elon Musk has published his LLM in open access — Grok even releases the information that is considered prohibited in ChatGPT. This is one of the largest neural networks in the world — it was trained on 314 billion parameters and weighs as much as 300 GB.

This typically doesn't matter too much when you're just generating funny elephant pictures. But if you're building consumer or professional apps, you don't want to leak sensitive and private data. Retrieval-augmented generation (RAG) compounds the problem, by feeding even more data into the model — silently, in the background.

Google renamed Bard to Gemini and is now rolling out a mobile app and Gemini Advanced with Ultra 1.0.

Dive into the world of data visualization with ChatGPT's expert guidance on D3.js bar charts. This tutorial simplifies complex concepts, ensuring anyone can master the art of creating interactive, data-driven charts. Perfect for beginners and pros alike!

Unveil the power of ChatGPT from OpenAI to revolutionize your coding workflow. Manual GIT commands? A thing of the past!

Optimize your React.js app with ChatGPT! Discover valuable tips and strategies in this video to boost performance effortlessly, with expert guidance from ChatGPT. 🚀

Unlock the potential of TypeScript with ChatGPT's expert assistance! In this video, explore how ChatGPT can seamlessly help you rewrite legacy JavaScript files and convert them into TypeScript, giving your projects a well-deserved boost.

An advanced no-code pipeline builder like Catalyst QuickML enables developers, regardless of their ML proficiency, to easily create end-to-end ML pipelines.

Just to keep you in the loop on current AI tools for Dev Productivity.

Discover how AI can be your ultimate ally when you're faced with perplexing bugs and unexpected issues in your code.

In this exciting episode of AI Cookbook, we're diving into the world of code refactoring with the power of ChatGPT.

GitHub Copilot Chat is here for all, not just business users!

Tired of retyping ChatGPT prompts? Free Chrome extension to the rescue.

Nice experience of building an AI-based project.

Let’s get serious. Are your users secured against AI hacking?

This episode is special because it's not so much about Svelte in particular, but relevant for developers in general. If you haven't gone down the AI rabbit hole yet this episode is for you. Swyx gives us a crash course in what it is and how you can use it as a web developer. Enjoy!

Looking for AI API? ChatGPT API vs Google’s Palm API

Google’s Bard learnt how to access Google search. Wanna use it? Google Bard “Extensions” Are Here

Wanna have you custom LLM running locally? How to run Llama 2 on anything

Anders Hejlsberg and Daniel Rosenwasser of TypeScript fame are just two of the prominent names attached to this project demonstrating the huge interest within MS for LLMs. TypeChat’s goal is to address the issue of LLMs outputting unstructured natural language and to direct output into a typed form.

Wanna train your local LLM on your data? Vectorize your code repositories and PDFs with AI

While it's less tied to JavaScript, as it uses WebGPU, it offers yet another way to run large language models directly within the browser, all controllable through JavaScript.