PatrickUllrich

@KlausCodes You able to DL more than 135 yet? 🥱

@ItsPatrickU Too old to ask as well pack it up unc

No idea what it takes to “distil a model” and too afraid to ask now

@thepanta82 I thought this was rage bait. This better be also a way better solution. Else this would take 5+ years to be worth it and have 0 maintainance needs. (Assuming dev makes at least $500/day which is low if it’s not an intern) lmao

We just replaced a $288/year SaaS with an internal solution. 2-3 days of work for 1 dev.

@devagrawal09 I m starting to think you almost need to write the contracts / api yourself and then let it fill in the blanks. Iterating on it is… okay, but the problem gets even worse once context grows even just a bit.

Man LLMs make the worst possible architecture decisions 80% of the time the best solution is so obvious and still it completely misses it and ends up building piles of shit Are there people that actually trust these models to make good decisions?

@addyosmani Myself to write more code again, at least on bigger features. Even if it’s just the contracts of functions, placeholders, or a general direction. On the flip side, it’s exciting to see that there are definitely a lot of “dark factory” opportunities today.

@addyosmani Different than a human would. It’s subtle. But after a few months it becomes systematic and noticeable. Maybe that’s fine, and maybe it’s the knowledge debt (not knowing where that function is, or you find this piece of code) that gives me the weird feeling, but I m finding…

@nater02 @expo Is there a repo to check out the code?

Flighty is one of the best-designed native app's on iOS, so I rebuilt it with @expo. Everything on the Expo side is real native UI. This is the best time there's ever been to build app's with Expo.

@BraedendotTECH Nah, same. Weird times. Also way more exhausted. Turns out reviewing code all day is way more straining than writing it.

I'm 33 and I think Claude Code is melting my brain. For 6 months straight I've had 5-6 terminals open at once, waiting on responses just to smash "enter" 90% of the time. That's the whole job now. And it's doing something to me. A few friends and I keep circling back to the same thing in conversations: none of us feel as sharp as we used to. Maybe it's just us. But I keep wondering how many other people in their 30s feel it too. (And yeah: this is a me problem, how I lean on the tool, not the tool itself. Doesn't make the effect any less real.)

@james406 Developers are spending up to 50k/mo on lava lamps I heard and leaderboards have popped up internally keeping track on who has the largest collection.

fun fact, cloudflare has these lava lamps at their headquarters up until today, nobody knew why i just learned that it's because the CEO misheard "lava lamps" as "LLMs" and accidentally ordered WAY too many, thinking they were ChatGPT licenses pretty neat!

@AnimexTwts What are the chances one punch man redeems itself ?

2027 is going to be insane

@BHolmesDev It def did work better with fable than it did with opus 4.8. But wasn’t all smooth with fable either

@BHolmesDev It still feels a lot slower than Claude - which is fine, I just do more things in parallel. I played with Theos Screenshot and converted it to a skill. Any other recommendations? Am I the only one that keeps running into issues? Using warp + omc fwiw

This is the biggest changelog to Hubble.md I've shipped. Part of it is the incredible community participation on big PRs (thanks everyone!), and the other part is Fable 5. Absolutely sold on Fable for discussion + delegation, and GPT 5.5 for implementation using subagents

@thekitze Not the best representation of what one would think is Canada. It be like going to Germany expecting beer, lederhosen, Oktoberfest, and moving to Frankfurt.

@thekitze I’d recommend checking out Vancouver, Montreal, Calgary independently first before picking Toronto. Without commenting what’s the best, but you are doing yourself a huge disservice not checking those out if you are thinking Toronto. Toronto (besides diversity) is actually …

The amount of people who live in Canada that reply to not move to Canada is hilarious

@sivalabs Yes

While doing Spec Driven Development, I find myself in a situation again and again. I create (or generate) a Spec, review it, if it looks good, I ask AI to implement it. Once AI is done, when I take a look at the code, I realize I missed a couple of small things. They are not too big to update the spec and ask AI to check for delta and patch the implementation. So, I quickly make changes, but am too lazy to update the spec knowing there is less likely I will take a look at the already implemented spec in future. When in doubt I check the code. Now there is drift in the spec and implementation. Is this what some people are referring to when they say SDD doesn't work because we can't write a complete spec ahead of time?

@mitsuhiko The worst part are comments that describe in session progress which makes no sense looking at the PR as a final state - referencing things it wrote and has since deleted 🤦‍♂️

Fable really likes its comments. Damn.

I have yet to hit a fable to opus downgrade… What am I doing wrong? 😢

@karrisaarinen @chiplay @linear Or is sentry the one you actually get your analytics from

@karrisaarinen @chiplay @linear This looks like what our step would look like, but datadog integration can only create linear tickets from what I can tell. Is there another one or is there a way to do a custom mcp for this?

A bug enters triage. Linear instantly labels it, runs a full code investigation, and opens a PR for your team to review. Automated triage that actually closes the loop. 🔁