AI/ML

• How to use GitHub Copilot: Prompts, tips, and use cases - The GitHub Blog Good advice how to write prompts that have sensible results.
• Neural Networks: Zero To Hero A course by Andrej Karpathy on building neural networks, from scratch, in code.
• Prompt Engineering vs. Blind Prompting – Mitchell Hashimoto A lot of people who claim to be doing prompt engineering today are creating prompts with a crude trial-and-error approach paired with minimal or no testing.
• Neural Networks: Zero To Hero A course by Andrej Karpathy on building neural networks, from scratch, in code.
• Accessibility and Artificial Intelligence - Joe Dolson Web Accessibility Automated captions are not enough and analysing an image using AI gives you a good description but not a good alternative text.
• Prompt Engineering | Lil’Log Detailed information about what PE is and what it is not.
• mlc-ai/web-llm: Bringing large-language models and chat to web browsers. Everything runs inside the browser with no server support WebLLM is a modular package that brings language model chats into web browsers with hardware acceleration.
• AI Playground AI Playground allows you to query several LLMs at once and see the difference in the results.
• suno-ai/bark Bark can generate highly realistic, multilingual speech as well as other audio - including music, background noise and simple sound effects.
• Unpredictable Black Boxes are Terrible Interfaces Why generative AI tools can be so difficult to use and how we might improve them
• Post-GPT Computing | Grady Simon Over time, we’ll see that what many of us really liked about building software deep down wasn’t coding, but intricately imagining the things computers could do for us and then getting them to do it.

• [How GitHub Copilot is getting better at understanding your code

  The GitHub Blog](https://github.blog/2023-05-17-how-github-copilot-is-getting-better-at-understanding-your-code/)With a new Fill-in-the-Middle paradigm, GitHub engineers improved the way GitHub Copilot contextualizes your code.

• nat/openplayground: An LLM playground you can run on your laptop An LLM playground you can run on your laptop.

• [Large Language Models are Human-Level Prompt Engineers

  OpenReview](https://openreview.net/forum?id=92gvk82DE-&ref=labnotes.org)We propose an algorithm for automatic instruction generation and selection for large language models with human level performance.

• TabbyML/tabby: Self-hosted AI coding assistantSelf-hosted AI coding assistant. An opensource / on-prem alternative to GitHub Copilot.
• reworkd/AgentGPT: 🤖 Assemble, configure, and deploy autonomous AI Agents in your browser.🤖 Assemble, configure, and deploy autonomous AI Agents in your browser. 🤖
• Neither artificial nor intelligent Hidde de Vries taking a look at the current state of AI and ML. It’s not as smart as we think it is.
• The AI Future of Frontend Development Hans Reinl with an super thorough article on the future of frontend development. AI will augment our work, not replace it. And there are some interesting new roles emerging.
• OpenAI Plugin API Learn how to build a plugin that allows ChatGPT to intelligently call your API.
• Using Github Copilot for unit testing CoPilot is pretty good in writing the repetitive and boring testing boiler plate for you, but there is still a danger that you’ll test the wrong things.
• Can the AI revolution at least write all my stupid unit tests? Chris Coyier venting on annoying unit test boilerplate.
• Voice spoofing 3 seconds of audio is needed to spoof a voice according to Microsoft’s AI. Already in use by scammers to simulate family members.
• The generative AI revolution has begun—how did we get here?
• Bing-Chat - a Node.js client for the unofficial Bing Chat API. It’s like ChatGPT on steroids 🔥 Using the Bing Chat API in Node.js
• Copilot Internals How does GitHub CoPilot work?
• Creating Slides using GPT Adding GPT to Google Slides via a plugin - not sure about that as to me slides are wallpaper for your talk and a means to structure myself
• Custom Diffusion How Adobe’s Custom Diffusion tries to beat Stable Diffusion
• Dune AI How Dune could look and what is the impact of AI matching styles of other directors
• LLM How do large language models like ChatGPT work?
• Microsoft unveils AI model that understands image content, solves visual puzzles Microsoft AI model to understand image context and solve visual puzzles - could that be used to crack CAPTCHAS?
• NVIDIA Broadcast 1.4 Adds Eye Contact and Vignette Effects With Virtual Background Enhancements NVIDIA’s software that makes it seem that you are keeping eye contact, whilst you don’t look directly at the camera.
• Scribbled diffusion Paint a scribble - get an AI generated image
• Stable attribution An attempt to give original artists credit in ML generated images. Kind of doomed to fail?
• Stable Diffusion 2.0 and the Importance of Negative Prompts for Good Results We need to teach the machines when they are wrong to improve results
• The Expanding Dark Forest and Generative AI Musings on the dangers of generated content
• What Is ChatGPT Doing … and Why Does It Work? In-depth analysis on how ChatGPT works
• Why AI messes up fingers Fingers everywhere!
• World Building With GPT Can ChatGPT build game worlds?
• AI Generated Ficticious Movie Posters The movies that could have been.