Accidental prompt injection against RAG applications
Link: Accidental prompt injection against RAG applications: "@deepfates on Twitter used the documentation for my LLM project as a demo for a RAG pipeline they were building... and this happened: Why did their chatbot demo suddenly decide …"
This article helped me understand a little more about RAG.
Communication Structures
Link: Communication Structures: "Communication Structures."
Insights about communication across layers in an organization. Pretty sensible advice.
Puts Debuggerer 1.0.0 Includes Invoked Class/Method Name
Link: Puts Debuggerer 1.0.0 Includes Invoked Class/Method Name: "code master oop ruby jruby java jee jse desktop web gui glimmer swt libui tk gtk rails"
Yes we all do it. Debugging with print statements. This ruby package makes our print statements more informative and findable. Check it out!
Omakub
Link: Omakub: "Turn a fresh Ubuntu installation into a fully-configured, beautiful, and modern web development system by running a single command."
I've been doing a lot of setting up lately. Between K3S, Docker, Robots and so on. I've become good friends with the shell, yaml files, bash scripts, and much more. This article shows a nice one-liner approach:
Doing Stuff with AI: Opinionated Midyear Edition
Link: Doing Stuff with AI: Opinionated Midyear Edition: "AI systems have gotten more capable and easier to use"
I think many technology savvy people are way underestimating the usefulness of large language models, LLMs, like ChatGPT, the best known LLM. But there are others. And they are improving and changing on a weekly basis. But my respect and admiration for their capabilities also makes me (helplessly) worry about how they WILL be used to cause serious harm in the future. The attached article is a great run down of current capabilities of these systems.
How to Use Tailwind CSS for Your Ruby On Rails Project
Link: How to Use Tailwind CSS for Your Ruby On Rails Project: "Let's see how we can use Tailwind CSS in a Rails application and explore its utility-first approach."
For a while now I’ve seen mentions of the Tailwind css library. I’ve been a Bootstrap guy forever. The attached article is a good explainer to learn about Tailwind and Rails. (The world of tech never stops even for a moment)
From the Circle to Epicycles (Part 1) - An animated introduction to Fourier Series
Link: From the Circle to Epicycles (Part 1) - An animated introduction to Fourier Series: "A visual introduction to Fourier Series"
An amazing tour de force in mathematical visualization. Honestly I can follow the first half and then I am lost. But I really appreciate the creative and beautiful animations.
Testing with Python (part 6): Fake it...
Link: Testing with Python (part 6): Fake it...: "Things might get harder, though"
This article is part of a series on … testing with python! It is an excellent overview of mocking, explaining the why, but more importantly showing lots of examples and assessing how well and conveniently they work. Worth reading.
Ideas are Dimes a Dozen: Large Language Models for Idea Generation in Innovation
Link: Ideas are Dimes a Dozen: Large Language Models for Idea Generation in Innovation: "Large language models (LLMs) such as OpenAI's GPT series have shown remarkable capabilities in generating fluent and coherent text in various domains. We compar"
I often say that ideas are cheap. And by that I mean that everyone has a million decent ideas. The art/trick/skill is to know how to turn that one idea into something that people want, use, benefit from, and maybe are willing to pay for. That's why I don't often worry about "who had that idea originally" (and that includes me) but am more impressed with "who was able to execute successfully around that idea". And so, I don't doubt (from personal experience) that LLMs now and more so every day, will accelerate finding, sorting through, testing and challenging, ideas that lead to startups that work. Here's a paper on that very question.
