Blog

Today we are bringing some unifying good news for all Racket and Chez Scheme folks having to run their applications on that alternative operating system called Windows. With Racket 8.17 released, it is possible to fully suport even mouse input in terminal applications on that strange platform!

Embedded platforms are ubiquitous, however they usually have limited resources. However a web interface is often useful for these systems. In this article we explore the possibility of embedding React app in statically-linked CHICKEN binary that can be deployed to such systems by simply copying it where it can be run as a system service.

Recently we had to get our last mile Internet connection upgraded. Switching to 60GHz technology gives us shorter rount-trip time and allows for higher bandwidth up to 1Gbps right now and about 2.5Gbps in the future. But being a wireless technology, some work had to be done physically on the roof of our offices building.

Last time we have shown how to run (or more like crawl) LLM inference on common CPUs. It can be rather sluggish though. And that is why we dug deeper and tried to use a commodity hardware - a gaming laptop - to speed things up! It is actually very interesting how even a gaming GPU can increase performance of AI tasks.

One of our current projects has to use LLM for extracting structured data from rather unstructured and noisy input. Such endeavor typically requires specialized GPU for decent inference times and it would be wise to test the solution with cheaper setup before buying such expensive hardware. It turns out it is possible to test even large models on CPU.