Change link to mellea

Author: robertmyslivecek

_posts/2025-08-14-thinking-about-ai.md

@@ -15,7 +15,7 @@ What’s the problem? In large part, I would argue that it has to do with anthro
 **In this Generative Computing blog series, we’ll explore the alternative framing that large language models are best thought of as computational engines, not so different from the computers that we’re used to today.** Like computers, they take instructions (albeit written in natural language), and they process various kinds of input data, and transform it into output data. This is not a new observation—and arguably part of what we’re doing here is to give name to a trend that is already emerging in the field. However, I would argue that we’ve only begun to scratch the surface of embracing generative AI as computing. I believe that taking this idea seriously will lead us to new programming models for interacting with LLMs, new tools and patterns for LLM usage, and even new ways of training LLMs. At the core of our philosophy is a belief that the full potential of generative AI will be realized by weaving AI together with traditional software in a seamless way. Generative computing describes a worldview where LLMs are an extension of computer science, not some alien entity set apart from it.[^1]
 
 We’ll use the term “generative computing” as an umbrella over a wide range of topics. Computer science is a broad field of study, covering everything from computing hardware, to software architecture, to human-computer interaction. We believe that generative computing has the potential to be similarly broad, and our posts will reflect this diversity of topics.
-Along the way, we will also introduce **Mellea**, an open-source python library where we are working to bring some of the key ideas of generative computing to life over time. If you’re eager to get your hands dirty right away, you can go straight to [**mellea.ai**](https://mellea.ai/overview/project-mellea) to get started. It’s early days, so expect some rough edges, but we’d love your feedback.
+Along the way, we will also introduce **Mellea**, an open-source python library where we are working to bring some of the key ideas of generative computing to life over time. If you’re eager to get your hands dirty right away, you can go straight to [**mellea.ai**](https://docs.mellea.ai/overview/project-mellea) to get started. It’s early days, so expect some rough edges, but we’d love your feedback.
 
 With our goals defined, let’s get started. In the next few posts, we’ll take a brief look at where the synthesis of generative AI and computing stands today, and we’ll lay out some of the challenges in current practice that led us to our current agenda. The purpose of these critiques is not to tear down or dismiss any of the amazing progress we’ve seen in AI. We’re optimistic about the potential and future of AI. Our goal is to identify problems so that we can fix them, and we’ll present one vision for what those fixes could look like, synthesizing some of our own ideas with a rich body of work from the field at large.