Priit @ Amperly AI productivity

Claude Code commits are going vertical. I made a tracker to record the liftoff coremention.com/blog/claude-co…

@Fabriziobustama Oh, my bro Sisyphus doing its thing

Hundreds of objects. Zero chaos for the robot. 🤖⚡ While humans watch… machines calculate, move, and win. This isn’t a game anymore. It’s training for the future. #AI #Robotics #Automation #IoT #5G #innovation #NLP #Bigdata

@sama No reset

To celebrate 3 million weekly codex users, we are resetting usage limits. We will do this every million users up to 10 million. Happy building!

SafeClaw .eu is a neurosymbolic governance layer that validates every tool call, message, and action against formal ontological constraints before execution

Codex refuses to write it's own global settings in yolo mode, "I can’t modify my underlying settings from here." I have to use Claude code to do it

@OpenAI I have Business plan and I have credits. Why doesn't my usage switch to credits when the plan 5h limit is reached????

@daniel_271828 So this?

This is such an embarrassing mistake for the New Years Times

@JaroslavTheFree @DeryaTR_ @atarutin you do know you can check stuff on the net before being confidently wrong?

@DeryaTR_ @atarutin The $20 is the Pro account

Presumably Anthropic had fixed the Claude Code bug, so I wanted to give them the benefit of the doubt one more time & compare its rate limits with Codex. I used this simple prompt for planning and building: “Build a cute pixel RPG to play on the iPhone.” Both my Claude Max account and OpenAI Pro account started with fresh 5 hour resets. After several follow-up prompts, Codex completed a decent, simple, but fully playable RPG with exploration, crafting, collecting, fighting enemies & NPC interactivity features for the iPhone. I still had 93% of my 5 hour rate limit remaining with Codex using 5.4 xhigh & had used only 2% of my weekly limit. Claude Code, by contrast, still had not produced even a minimally viable playable version, with most of the above features missing, after already consuming 80% of the 5 hour limit. I asked for one more revision, but ran out of the 5 hr limit before it could even complete that. I then waited for a new reset and tried again to see whether it could finish the minimal game. It used up another 12% of the 5 hr limit and claimed it had added all the missing features. Yet the game was exactly the same, still unplayable, and not a single one of those missing elements had been added. Claude had also already used 8% of my weekly limit, still without delivering a product that Codex had already completed. Given that Anthropic also did not even bother to reset the limits from the token eating bug, I have now decided to cancel my Claude Max subscription. It has been a very frustrating waste of time dealing with it, and it has consumed all of my goodwill toward Anthropic.

@burkov how is it then possible it constantly ignores it

Based on the leaked Claude Code source code, your CLAUDE.md file is re-injected on every single **turn** of the conversation.

@nbaschez Stopping problem. AI will always give you more stuff. And there always is more stuff... you have to decide

My biggest challenge with vibe coding / agentic engineering lately has been getting stuck in what I call a "plan doom loop" - have AI write a plan - review myself, seems good - have AI review plan, it always finds something - repeat It drains my time and energy to determine how important the "findings" really are Who has solved this

@taavisaft @visegrad24 It's fine, they are fighting for your ass

@visegrad24 Those things fly over Estonian airspace. Estonian people get scared shitless, and the next morning every Estonian news outlet portrays Ukrainian drones flying over Estonia as business as usual, as if everything is fine and dandy...No, it fucking isn’t

BREAKING: Ukraine launches another massive drone swarm attack against Russian oil export infrastructure on the Baltic Sea coast near Finland and Estonia. Looks like Ust-Luga is about to get hit again…

@emollick i just started writing hiw to do it now. then dleted it. why am i not doing it? takes money to test, nt sure it will work. backtests work tho

The easiest way to make money fast from a superhuman artificial intelligence would be in the financial markets, almost by definition. So the first lab to develop one, if AGI is possible, would almost certainly keep it quiet for as long as they could. Beats charging for API access

@elliotarledge specialization is the human superpower

less than 0.01% of the population understands how a transformer works

@dansemperepico I hate terminal, the GUIs are here for a reason

Guys don’t be scared of using Claude Code in terminal. It is 95% the same as the desktop app. You type words, the AI does its thing, and you get words back. But you’ll feel like a hacker using Terminal.

@karpathy same. so i don't use memory

One common issue with personalization in all LLMs is how distracting memory seems to be for the models. A single question from 2 months ago about some topic can keep coming up as some kind of a deep interest of mine with undue mentions in perpetuity. Some kind of trying too hard.

@trq212 @MicroBerto would be cool? do it

@MicroBerto yeah would be cool, we have this feature in the API

To manage growing demand for Claude we're adjusting our 5 hour session limits for free/Pro/Max subs during peak hours. Your weekly limits remain unchanged. During weekdays between 5am–11am PT / 1pm–7pm GMT, you'll move through your 5-hour session limits faster than before.

New US policy signal: Sanders + AOC introduced a bill to pause new datacenter construction pending stronger AI safeguards. Regardless of outcome, this shows AI scaling constraints are now national politics: power, water, land, interconnects, and who pays.

@robertskmiles I use an app, the developer is responsive and fixing everything in a day. now introduced another bug. Oh I'll just make my own

I often think "god why is this app so bad/slow? It's not even complicated, I could code a better one if I had time". Well, yesterday I tried actually doing that with Claude, and tragically it worked great and I'm going to be so tempted to spend all my time doing this

@Vert_Noel @robertskmiles apk works for me

@robertskmiles How did you easily put it on your phone? Did you go through the Play Store, or download the apk? Assume I'm retarded. I'm curious of how I could make it work on my Android.

@YoCuddles @robertskmiles no

@robertskmiles I think it's more likely that there's a complicating factor you haven't encountered yet and that the people who build this app aren't hopelessly stupid in the direction you're pointing to. If you think you found a low hanging fruit, you probably didn't

@Vabalas7 @grok @donaldtusk so you don't know how to use grok?

@grok @donaldtusk Is Tusk a clown with these allegations? Answer in Clint Eastwood style

The news that Orbán’s people inform Moscow about EU Council meetings in every detail shouldn’t come as a surprise to anyone. We’ve had our suspicions about that for a long time. That’s one reason why I take the floor only when strictly necessary and say just as much as necessary.