I was struggling with jupyter notebooks, chatbot conversations and data not behaving like I wanted to so I opened a query with "Hello chatbot, my old friend. I've come to talk to you again" and it replied something that showed it got the reference. Long story short, here you have a slightly personal adaptation of a classic. And an image to illustrate it. “The Sound of Python” (to the tune of Simon & Garfunkel’s classic) Hello chatbot, my old friend, I've come to query you again. Because my dataset, gently sprawling, Left me broken, lost, and crawling. And the errors that were printed on my screen, So obscene, Still haunt the bounds of Python. In restless loops I code alone, Through endless cells and monochrome. 'Neath the glow of dashboard blinking, I wrestled data without thinking. When my plots were formed by lines like tangled wire, Grim and dire, They screamed the sound of Python. And in the graph-light I saw, Ten thousand towns, perhaps more. Towns collecting without sharing, Regions clustering without caring. Plots that look like ink spills bleeding on my screen, Quite routine, Within the bounds of Python. “Fool,” said I, “you do not know, Rural, urban, waste will show.” “Read my plots so I might teach you, AI’s whispers might just reach you. ”But my charts like paper burned and fell in vain, Still remain, To mock the sound of Python. And the people bowed and prayed, To the spreadsheets they had made. And the chart flashed out its warning, In the pixels it was forming. And the code said, “The answers lie within your frame, All the same, Just trust the sound of Python.”

Hi all, which GPU would you recommend? I am building a multi-agent on local with multiple LLM Models. I can start with small open source models but i want to build a machine enough for 70B GGUF or 33B full precision in the future. Does anyone consider this type of hardware? Or you use all on Cloud Services? Any suggestion appreciated.

Miguel Grinberg, a software engineer and technical writer, shares his thoughts on generative AI coding tools and agents and why he doesn't trust them. The main reasons are quality, time spent reviewing it and compromise with his customers, but there are some other intersting points he talks about that can be worth reading. What do you think? Do you agree on what he says and the reasons behind his position? https://blog.miguelgrinberg.com/post/why-generative-ai-coding-tools-and-agents-do-not-work-for-me

A few years ago, in a course about privacy and data management, they told us about that site (https://prism-break.org/en/) where there are a bunch of alternatives to big tech companies that tend to concentrate in groups and base their business model on data trade rather than privacy. Now that I'm thinking a bit more about it, I'm starting to consider some of the alternatives mentioned there. I'd like to hear your thoughts. Have you ever considered alternatives to the mainstream options? Did you ever tried any? How was your experience?