Uwe Friedrichsen

@ufried.bsky.social

Uwe Friedrichsen. Longtime traveler in the world of IT. Dot Connector. Keeper of timeless wisdom. Translator between floors. System design. Resilience. Sustainability. Tries to make (not only) IT a (bit) better place.

every time, i open any social media feed (which i do increasingly infrequently), "everybody screams" to the tune of "everybody hurts" by r.e.m. starts playing in my mind ... 🤷

You can find the (unabridged) slide deck of my JAX 2026 talk "The 7 pitfalls of AI" at speakerdeck.com/ufried/the-7.... The deck discusses some questions, we still need to answer if we want to use AI at scale in a sustainable manner. enjoy if you like ... ;)

The 7 pitfalls of AI

This slide deck discusses some of the open question that the massive AI push still left unanswered. It starts with a little glimpse at the market mechan…

speakerdeck.com

the interview was in november which means, i do not have the slightest idea anymore what i said back then ... :D i hope it makes sense ... at least ... well ... somehow ... :) anyway, enjoy if you like ... ;)

Software Architecture Gathering@sagconference.bsky.social · 3mo ago

𝗧𝗵𝗲 𝗩𝗮𝗹𝘂𝗲 𝗼𝗳 𝗦𝗶𝗺𝗽𝗹𝗶𝗰𝗶𝘁𝘆 𝗶𝗻 𝗦𝗼𝗳𝘁𝘄𝗮𝗿𝗲 𝗔𝗿𝗰𝗵𝗶𝘁𝗲𝗰𝘁𝘂𝗿𝗲 ✨️️ 𝗘𝘅𝗽𝗲𝗿𝘁 𝗜𝗻𝘁𝗲𝗿𝘃𝗶𝗲𝘄 𝘄𝗶𝘁𝗵 𝗨𝘄𝗲 𝗙𝗿𝗶𝗲𝗱𝗿𝗶𝗰𝗵𝘀𝗲𝗻 💡In this #SAGconf interview with @ufried.bsky.social you'll learn: • why simple solutions are harder • why business understanding matters • how simplicity improves robustness 👉️️ t1p.de/iqu81 #SAG2025 #iSAQB

𝗧𝗵𝗲 𝗩𝗮𝗹𝘂𝗲 𝗼𝗳 𝗦𝗶𝗺𝗽𝗹𝗶𝗰𝗶𝘁𝘆 𝗶𝗻 𝗦𝗼𝗳𝘁𝘄𝗮𝗿𝗲 𝗔𝗿𝗰𝗵𝗶𝘁𝗲𝗰𝘁𝘂𝗿𝗲 ✨️ 𝗘𝘅𝗽𝗲𝗿𝘁 𝗜𝗻𝘁𝗲𝗿𝘃𝗶𝗲𝘄 𝘄𝗶𝘁𝗵 𝗨𝘄𝗲 𝗙𝗿𝗶𝗲𝗱𝗿𝗶𝗰𝗵𝘀𝗲𝗻

companies talk a lot about "generalists" and "polymaths" in the context of ai but their job ads still reflect the opposite. and according to some friends who just changed jobs, they definitely filter out anyone with a hint of generalist during their interviews. /1

is it just me or do you also get the "oops. something went wrong" page at linkedin regularly if you try to do something that is not just reading your feed? at least for me it feels as if linkedin's system quality went significantly down in the recent weeks.

just read in a client document: "with ai, we have to focus on value, e.g., features delivered". well, a feature delivered is _not_ value. it is just output. it only becomes value if the user accepts it (and is willing to use/pay for it). 1/2