A cognitive design I’ve come across recently apes the subconscious messages we have in our own brains, as distinct from inner monologue or stuff we say. Referencing our own way of thinking has revealed to me insight about how to improve GenAI functionality, as well as revealing back to me new insights into how we ourselves think. I'm a distinct amateur in neuroscience, so I hope if I blog this someone more informed could perhaps comment on the approach outlined below, but I am finding it a very useful technique.

See this post about how I define Cognitive Design for GenAI architecture

For this explanation, I break down cognition messaging into three modes:

I believe the messages passed around within a cognitive design can be broken out into the sub-categories above, and that can help us design better performing systems. This seems to become important once one starts to work with asynchronous, parallel calls to GenAI models, which again I think may be because that is more akin to how human brains work, as opposed to sequential, one call at a time API requests we start with when first getting to know GenAI models.

Every few years I feel the need to change my blogging platform, and each time I am compelled to write a blog post about the exciting new blog tech. I've moved through Blogpost, Wordpress, Posterous, Jenkins, Hugo and today I'd like to introduce Docusaurus.

And since this is a GenAI blog, it makes sense I selected a new blogging platform I feel will support GenAI. Its a little thought provoking that the current GenAI models work best when working with the most popular languages, frameworks or opinions. They are after all approximating the average of all of human expression. This means they will do better at English, Python and React than more niche areas such as Danish, R or Vue. I hope this does not destroy diversity.

But it also means that since it seems React is the most popular web frontend framework at the moment, it makes sense to investigate using React within GenAI applications.

This Docusaurus blog is written in a flavour of Markdown that supports React Components which made me think: is this a good vessel for creating GenAI output that can dynamically adjust its output format? Can we go beyond text to dynamic user experiences depending on what they need? Lets find out.

Its more than a year since my blog post about running LLMs on Google Cloud was published, and its safe to say it changed my life. It feels appropriate to publish a follow up here within the sunholo documentation, as its where I've been mentally living for a lot of the past 12 months. The reaction to the above blog encouraged me to change my career direction and embrace GenAIOps within my own startup, Holosun ApS. This year has been one of intense learning and excitement with ups and downs, but I feel like I made the right call.

This blog post will take some of what I've learnt this past year within GenAI and apply it to a common task in my pre-GenAI career: what is the best way to use BigQuery to extract insights from Google Analytics 4's BigQuery data export? I wrote a book on Learning Google Analytics in my career before GenAI - with these amazing new tools, can we get more value than before? To tackle this in a GenAI way, let us meet Bertha.