AI is Getting Smarter Every Month… Are We So?

That’s right, two new apps in one week! I have been working on two new apps, Squarish.app and AgentUML.com, and I’m excited to share them with you. Both apps are designed to enhance productivity and streamline workflows for developers and AI enthusiasts.

Squarish.app is a tool that allows you to create and manage DIY pixel art projects. It’s perfect for artists, designers, and hobbyists who want to create pixel art without the hassle of complex software. With Squarish.app, you can easily design, edit, and share your pixel painting art creations.

AgentUML.com, on the other hand, is a nice little tool that allows you to generate UML diagrams from text chats with AI assistants. This app is ideal for developers and software engineers who want to quickly visualize their code structure and design patterns without manually drawing diagrams. By simply describing your classes, relationships, and interactions in a chat format, AgentUML.com will generate clear and accurate UML diagrams for you. The nicer part of it is that it can help LLMs (Large Language Models) to understand the code structure better and generate more accurate code suggestions. So to use it in code generation, you can describe your code structure in a chat with an AI assistant, and then use AgentUML.com to generate the corresponding UML diagrams. This can help you visualize your code and improve your understanding of its architecture. PRDs (Product Requirement Documents) can be hot to describe long running agent tasks, and AgentUML.com can help you help the LLMs to understand the code structure better and generate more accurate code.

Both apps are built with the latest web technologies and are designed to be user-friendly and intuitive. I believe they will be valuable tools for anyone looking to enhance their productivity and creativity in the digital space. Right now, both apps are in their early stages, and depending on the feedback I receive, I may add more features and improvements in the future.

AI Assistants and The Learning Curve

AI is a child in human hands, and it is still learning, and we are still learning how to teach it and how to use it effectively. The learning curve for AI assistants can be steep, but with the right tools and resources, we can make it easier for developers and users to harness the power of AI. Both Squarish.app and AgentUML.com are designed to help users navigate this learning curve by providing intuitive interfaces and clear instructions. But that is still only accessible to those who are willing to learn and experiment with AI.

I’m a proud Twitch.tv watcher, not a streamer, and I find it fascinating to watch how people are using techlogy to broadcast their lives and share their knowledge all through the internet in their home spaces. I have been watching a lot of Twitch streams lately, and I have noticed that many streamers are still hand coding their apps and software, which can be teadious to watch as a AI coding assistant user.

My thoughts lately have been about how to make it easier for developers to use AI assistants in their coding workflows. Still yet, I would like to teach and guide them on how to use AI assistants effectively, and I believe that Squarish.app and AgentUML.com can be valuable tools in this regard. And not only that, but I also want to make it easier for developers to use AI assistants in their coding workflows. By providing clear instructions in an arborescent development schema-from libraries to frameworks to applications to modules to components to classes to methods and functions to properties and constants, I hope to make it easier for developers to understand how to use AI assistants effectively in their coding workflows. I don’t even use a code editor window anymore, I just use a terminal window and a browser window for web apps, a git client, and an AI assistant chat window. I have been using this setup for a while now, and I find it to be very efficient and effective. I can quickly switch between different tools and resources, and I can easily access the information I need to complete my tasks. All that through language agnosticism.

R2-D2 In The Next Fifty Years

I have been thinking a lot about the future of AI and how it will impact our lives in the next fifty years. I believe that we will see significant advancements in AI technology, and we will likely see AI assistants become more integrated into our daily lives. I envision a future where AI assistants are as common as smartphones, and they will be able to assist us with a wide range of tasks, from managing our schedules to helping us with our work and personal projects.

I also believe that we will see the development of more advanced AI assistants that can understand and respond to human emotions and needs. These AI assistants will be able to provide personalized recommendations and support, and they will be able to learn from our interactions with them to improve their performance over time. In this future, I imagine a world where AI assistants are like R2-D2 from Star Wars, always by our side, ready to help us with whatever we need.

But There’s A Bad Side To This

While the future of AI holds great promise, there are also potential risks and challenges that we must consider. As AI assistants become more integrated into our lives, we must be mindful of issues such as privacy, security, and ethical considerations. We must ensure that AI technology is developed and used responsibly, and that it benefits society as a whole. Additionally, we must be aware of the potential for job displacement and the impact that AI may have on the workforce. It is important to strike a balance between embracing the benefits of AI and addressing the challenges that come with it.

While I don’t think jobs will be losts but rather transformed, I do think that the way we work and the skills we need will change significantly. We must be proactive in preparing for these changes and ensuring that individuals have the necessary skills and resources to thrive in an AI-driven world. That is from vision AI, where AI can see and understand the world around it, to language AI, where AI can understand and generate human language, to reasoning AI, where AI can make decisions and solve problems based on complex data and information. But for the reasoning part I am still skeptical about how much reasoning AI can do. We still have much of the solar system to explore, not to mention the galaxy. What we’ll find in those realms will be beyond our imagination, and AI will play a crucial role in helping us navigate and understand these new frontiers.

A Deductive Model, Not Predictable

I have this idea of a deductive model for AI, where AI can learn and reason based on a set of rules and principles. This model would allow AI to make decisions and solve problems in a more univeral (whatever of the universe we know until now) way, rather than relying solely on data and patterns. By using a deductive model, AI could potentially reason and make decisions in a more human-like manner, taking into account context, logic, and ethical considerations. This could lead to more advanced AI systems that are capable of understanding complex situations and making informed decisions.

It’s just an idea…