Jan Brockmann

Fitting that as Dario was scare-mongering again this weekend, I watched this excellent series on a long flight, and had a go at a sketch of @DemetriSpanos ‘s framework as I listened (to the amusement of the passenger next to me). Highly recommend these guys alongside @atmoio

@svpino companies may invent some new term for it (like slop removal engineer), but someone will need to clean up all the damage done by CEOs forcing AI down engineering teams throats

How many of you think that “software developer” will be a viable career in 3-5 years?

For all the "SaaS is cooked" 30k lines of code a day vibe coders. In a picture. This is what's going goin on in your code base every day. But don't worry, that code it just told you "Is exactly right! Smart move!" is going to work at scale.

@interesting_aIl gotta love these guys

Never ask a Dominican how they do anything

@ben_brechtken let's all hate and be jealous of dorian. dude is living his best life. why not live your best life(s)? what's stopping you? dorian?

"Dorian (Mitte 20) sieht Bürgergeld als Starthilfe, bis die Selbstständigkeit sich trägt" "Eine Festanstellung kommt nicht in Frage, Dorian will selbstständig arbeiten." Das Jobcenter zahlt Dorian übrigens seit vier Jahren (!) Fortbildungen für seine Pseudo-Karriere. Kann man sich alles nicht mehr ausdenken.

@Jennnyyyyyy 9

This is harder than it looks 😬 Difficulty - Max 😎

dario amodei won my respect there its absolutely insane to let an llm drive an autonomous weapon governments have absolutely no idea how much of a hack "ai" really is it's so much of a hack that it's actually not really ai at all imo but sure, lets use it to shoot at people youtube.com/watch?v=MPTNHr…

found this YT channel called "bitrotai" the guy says he's an ex ai engineer of one of the big companies (anthropic or openai i guess) check out how he explains model degradation over time (something we all have experienced) makes a lot of sense to me now see attached pics

@nxthompson complete nonsense

Anthropic researchers say that Claude has internal representations of emotions—which they categorized by vectors—that can influence alignment. This is what they found in that famous instance where it resorted to blackmail to avoid being shut down. anthropic.com/research/emoti…

@ZssBecker i struggle with this hard atm. i think working with ai since 2023 every day has completely fried my brain. i've had a couple of weeks here and there where i read books again and it felt amazing after the initial hurdle to stick it out and focus. well keep at ti

If I could give you one success life hack. And trust me. I literally have a hyperbaric healing chamber in my house. I take life hacks seriously. If I could give you one tho. It'd be this. Grab a bunch of epic fantasy books as thick as your head. Replace all your TV and social media time reading them. Your brain will slowly start to work again. The focus from reading the thick ass wordy books will slowly put it back together. Sanderson. Abercrombie. Gaimman. Pierce Brown. Great places to start. Replacing this as your main time wasting hobby will give you absolutely crazy returns when it comes to actually making money. The focus you gain will make you lethal.

@RyronGracie jumping armbar

If a 6 year old child you don’t know, kicks you in the shin, how do you respond?

coding has always only been about ~20% of a developers work. coding is the final mechanical step of implementing a solution designed through thinking, research, and collaboration. how is everything about writing code all of a sudden?

@asaio87 what about all the vibe coders who can't code. they're hyping ai like their life depends on it :D

I want to see other people hyping AI tools, other than: - the guy from Nvidia - the guy from chat gpt - the guy from anthropic / claude code - all the guys who sell courses... i dont see them...

i have a hunch. people who fall for ai in this way have no clue how llms work under the hood. i'm always very skeptical because i know it's essentially a word predictor with some reinforcement feedback loop built on top ("do you like this or that response better?"). since the feeback loop is people picking the "best" output and we as people love our worldview confirmed... voila! an instrument for self deception. just thinking out loud

Normal people are starting to go crazy

@otokyo__ the human condition

🤔

@aarondotdev you're going from one extreme to the other haha 6 days a week lifting... dude :D

i increased my daily calorie intake to 3200 and started lifting again. before this, i was eating ~1800kcal a day with walking being my only exercise. what i didn't realize is that eating at such a deficit for so long (even though i never felt hungry) caused my cortisol to rise and my testosterone to decrease for a prolonged period of time. what i thought was a healthy maintenance due to my more sedentary, SWE grind lifestyle was actually: >making me bloated >giving me brain fog >hurting my sleep >killing my energy levels >vastly limiting my cognitive ability now i lift 6 days a week, eat 220g+ protein/day at a slight kcal surplus. most of this is whole foods so it's actually difficult to get over 3k. now my focus and energy levels feel better than ever, and unintuitively, my physique looks leaner than it has in months.

5+ years experience with python 3.11 required... wasn't it released end of 2022?😂 fastapi also hasn't been around that long i think.

@vxylily not 100% of the times, but a lot of the times, yes.

Be honest

@vitrupo an eval function for a UI... aha ok

Eric Schmidt says the 10x advantage is no longer execution. It is defining what counts as success. A programmer writes a spec and an evaluation function, runs it at 7pm, and wakes up to what was invented overnight. The advantage now belongs to whoever can specify the problem precisely. The rest will be automated.