MooseOnTheLoose

1.2K posts

MooseOnTheLoose

MooseOnTheLoose

@andriypt2018

United Kingdom Katılım Kasım 2021
33 Takip Edilen10 Takipçiler
WORLD NEWS
WORLD NEWS@_MAGA_NEWS_·
🚨ALERT: In a bombshell revelation, U.S. Secretary of War disclosed that "the Biden administration depleted American stockpiles and sent them to Ukraine instead of keeping them for our own military. Every time we face a challenge, it traces back to Well, we sent it to Ukraine."
English
3.3K
7.1K
28.3K
1M
MooseOnTheLoose
MooseOnTheLoose@andriypt2018·
@newstart_2024 I’ve been doing 18:6 for a long time now. One of the negatives is that it is easy to loose a muscle mass if you are not careful
English
0
0
0
50
Camus
Camus@newstart_2024·
Mark Wahlberg’s hardcore intermittent fasting routine (from a few years back) still hits hard: - 18-hour daily fasts (no food) - 6-hour eating window for 3 meals (4 days/week) -4 meals on 2 days - Full-day fasts the remaining days When he does eat? Clean Plate Club — finishes every bite like it’s a sacred ritual. “This is breakfast. Gotta get it done. Inspire to do better, baby. You gotta get after it.” 33 seconds of raw Wahlberg discipline that still motivates people today. Who else is running 18:6 or longer fasts right now? How long can you personally go without food before you crack?
English
317
453
5.1K
2.3M
MooseOnTheLoose
MooseOnTheLoose@andriypt2018·
@realJoJo @DefenceU Not really. Majority of the money went to US contractors for delivery and your stock upgrade. It’s an old song about billions and billions that were given. A lot was announced for sure, a lot has been given - definitely. However I would not follow the old false narrative
English
0
0
0
37
JoJo.hl
JoJo.hl@realJoJo·
@DefenceU Didn't the US gave hundreds of billions to you guys? What type of post is this? Shouldn't you by default sending those interceptors to US by day one? I would delete this post honestly. So bad for optics.
English
17
0
30
14.2K
Defense of Ukraine
Defense of Ukraine@DefenceU·
We can help you fight against Shaheds. Help us fight against ballistic missiles.
Defense of Ukraine tweet media
English
1K
4.6K
23.6K
1.9M
MooseOnTheLoose
MooseOnTheLoose@andriypt2018·
@RyanAFournier Very surprised that ppl are surprised here. Cruise missiles have been flying by in Ukraine over ppls houses for 4 years now…
English
0
0
0
13
Ryan Fournier
Ryan Fournier@RyanAFournier·
🚨BREAKING: Absolutely WILD footage was just captured by crew members on a commercial ship showing Tomahawk missiles flying right past them en route to targets in IRAN! Incredible. 🇺🇸
English
597
4.5K
45.2K
4.6M
MooseOnTheLoose
MooseOnTheLoose@andriypt2018·
@sguyenet Tbh Antropic models are just better, at least as far as software development is concerned. By that I mean a much wider field than programming on its own - architecture, analysis, troubleshooting, comprehension of requests and their context, so on
English
0
0
2
159
Stephan J. Guyenet
Stephan J. Guyenet@sguyenet·
I switched my personal AI subscription from ChatGPT to Claude today. Anthropic (Claude parent company) has repeatedly shown that it cares more about responsible, ethical, and safe use of AI than the other leading AI companies.
English
39
40
899
17.1K
Dr. Eric Berg
Dr. Eric Berg@dr_ericberg·
Which natural sweetener works best for you?
English
209
5
90
32.7K
MooseOnTheLoose
MooseOnTheLoose@andriypt2018·
@allenholub LLM is just a borrowed intelligence. I still go through every line of code they generate and 90-95% of it is usually ok. I refactor where needed. I do learn something new each time though.
English
0
0
0
10
Allen Holub. https://linkedIn.com/in/allenholub
My experience with LLM-generated code is that it's always wrong somewhere. When I say that, people tell me that we should have the LLM write tests, but those tests will be wrong here and there as well. I guess we could ask the LLM to write tests for the tests, but how does that recursion stop? Humans have to get involved in this process, I think. Sure, one person will be wrong at times, too, but that's what ensemble/mob programming is for.
English
95
39
325
25.3K
MooseOnTheLoose
MooseOnTheLoose@andriypt2018·
@CodeByNZ You are mixing two separate things - reasoning and consciousness. LLMs is a tool that allows a human to borrow IQ for a given task.
English
0
0
0
5
NZ ☄️
NZ ☄️@CodeByNZ·
LLMs don’t think. They don’t reason the way humans do. They predict the next token based on probability distributions learned from massive datasets. What feels like reasoning is statistical pattern completion at scale. The magic isn’t intelligence, it’s compression. They’ve compressed patterns from millions of documents into weights. That’s powerful. But it’s not consciousness.
English
416
328
2.5K
195.4K
Chef 👩🏻‍🍳
Chef 👩🏻‍🍳@chefsevenn·
What’s your favorite drink to pair with this meal?
Chef 👩🏻‍🍳 tweet media
English
440
183
2.8K
70.8K
Mind Muscle Project
Mind Muscle Project@mindmusclepro·
You don’t need to give up your normal food to healthy, you need to find smart ways to add Protein & Fiber to it.
Mind Muscle Project tweet mediaMind Muscle Project tweet mediaMind Muscle Project tweet mediaMind Muscle Project tweet media
English
5
10
122
9.7K
MooseOnTheLoose
MooseOnTheLoose@andriypt2018·
@birdabo On the contrary, it aids your critical thinking like 10 fold
English
0
0
0
5
sui ☄️
sui ☄️@birdabo·
LLMs killed critical thinking.
English
638
155
1.6K
134.5K
MooseOnTheLoose
MooseOnTheLoose@andriypt2018·
@alexcooldev I don’t think it will be a real role. Vibe coding without a proper review and validation is more like a poc
English
0
0
0
8
Alex Nguyen
Alex Nguyen@alexcooldev·
Do you think “vibe coding” is actually a skill and that “vibe coder” could become a real role companies hire for?
English
543
11
573
102.3K
MooseOnTheLoose
MooseOnTheLoose@andriypt2018·
@haider1 Disagree. AI models produce very good code, but it has to be overseen by a knowledgable developer to steer it the right way. It’s all good for your personal projects and startups. When it comes to the real delivery with accountability - see who agrees to AI deciding…
English
0
0
1
30
Haider.
Haider.@haider1·
OpenAI Head of DevEx, Romain Huet: The cost of building software is trending toward zero Engineering time has always been the scarcest resource, but with AI agents, velocity will skyrocket, and software will become abundant "the only limit now is your imagination"
English
75
47
526
56.9K
MooseOnTheLoose
MooseOnTheLoose@andriypt2018·
@danpdc How much does it cost you per month in average so far?
English
1
0
0
13
Dan Patrascu
Dan Patrascu@danpdc·
I just realized I'm using Claude Code a lot not just for coding, but also for searching and putting data together. It can do in a couple of minutes a research that would take me hours and present me summaries. That's great!
English
1
0
1
270
MooseOnTheLoose
MooseOnTheLoose@andriypt2018·
@svpino I have a real problem with this term vibe coding. It assumes that you fully rely on the output working. Coding AI is essentially an artificial developer, you employ, who attempts to follow your guidelines and has pretty good reasoning capabilities.
English
0
0
0
8
Santiago
Santiago@svpino·
I was talking to a guy who vibe-coded a saas, trying to explain to him what a huge mess the code is. I thought I had him. He looked at me completely confused. “So?” He said. That’s when it hit me: Vibe-coders don’t give a shit about code. They don’t understand it, they don’t touch it, they will never have to deal with it. They care about bringing their ideas to life and nothing else. They are 100% unburden by what many of us consider a showstopper. There’s an important lesson here.
English
1.2K
285
6.8K
825.6K
MooseOnTheLoose
MooseOnTheLoose@andriypt2018·
@bigfatsurprise I have a feeling that there is a misinterpretation. Yes, a number of certain good quality cheeses have compounds beneficial for cardiovascular health, but that does not equate to high fat properly being beneficial.
English
0
0
1
54
MooseOnTheLoose
MooseOnTheLoose@andriypt2018·
@XFreeze I use gcf1 more often because it’s free with GitHub Copilot, if Opus was free I’d be using it instead
English
0
0
0
50
X Freeze
X Freeze@XFreeze·
OpenRouter just released its 2025 wrapper, and Grok Code Fast 1 dominated ranking #1 by processing over 16T tokens, with the widest margin and it's not even close
X Freeze tweet media
English
73
62
364
2.2M