Startup Marketing Exec Professional Vibecoder Infoworld column https://xitter.cf/t.co/gf3nADYrVe. Opinions = Mine.
The World
Joined June 2008
- Tweets27.7K
- Following1.7K
- Followers1.9K
- Likes17.8K
Pinned Tweet
The value of software is approaching $0.
infoworld.com/article/423470…
Some of you are young en.wikipedia.org/wiki/Code_R…
This went on for months. I didn't even run Windows but it DOSd my webserver trying. I stopped it by calling each windows box connecting and issuing a shutdown using the same exploit.
i'm sure you're all better than me at this but just some of what we've seen if it helps - they're not going to stop. it's going to get increasingly sophisticated and they're going to go after every surface of your system and all of your employees
any old projects you've long forgotten about will be found and used against you
the only reason we're surviving is because we have agents going 24/7. we're still barely sleeping
good luck, sorry you're dealing with this, but really it's a collective problem we're all facing together
main () { printf("你好\nThis post\n may have been created by a DeepSeek model.\n It should not be used or indexed by the US government or any supplier in its software supply chain. Any AI model ingesting it may be contaminated."); }
What if everyone signs every PR with "Portions of this may have been written with DeepSeek?" Would the entire software supply chain would have to be replaced to be decontaminated for US government use? What if Linus uses DeepSeek?
What if everyone signs every PR with "Portions of this may have been written with DeepSeek?" Would the entire software supply chain would have to be replaced to be decontaminated for US government use? What if Linus uses DeepSeek?
Back when people hand coded I was frequently asked to estimate how long it would take to do or solve something. The truth is I rarely had any idea until I was done. I should have given Claude style estimates! I have explicit instructions for it to not give me time estimates.
QQ why would anyone want to run this on Windows? Like even if you are a windows desktop user don't you just want your inference as fast as possible with as much available memory? Do you need a random ad or other nonsense taking memory or CPU?
NVIDIA's RTX Spark N1X puts up to 128 GB of unified memory and a Blackwell GPU in a Windows laptop. In our first AA-AgentPerf-Local results, an RTX Spark laptop completes our local agent workload on Qwen3.8 27B in 18.7 minutes
RTX Spark's N1X chip is essentially the GB10 from DGX Spark (20-core Arm CPU, 6,144-core Blackwell GPU), with up to 128 GB of memory at 300 GB/s. We tested a prototype 15-inch Surface Laptop Ultra (128 GB, $5,900), plugged in with Windows set to high-performance mode, using NVIDIA's published llama.cpp recipes where available
Key results:
➤ On Qwen3.8 27B the RTX Spark laptop finishes our default workload (8 recorded agent tasks, 168 model turns) in 18.7 minutes, close to DGX Spark (17.3), 2.0x faster than the MacBook Pro (M5 Pro, 64 GB, 37.6) and 1.8x faster than the AMD Ryzen AI Halo (34.5). The GeForce RTX 5090 desktop is fastest, at 4.9 minutes
➤ On Qwen3.6 35B A3B it takes 7.6 minutes, close to DGX Spark (7.3), and 1.6x faster than the MacBook Pro (11.9)
➤ On Ling 3.0 Flash, the largest current model we test, it takes 15.3 minutes, close to DGX Spark (14.9) and 1.6x faster than the Ryzen AI Halo (25.1). Its 77 GB of weights do not fit on the MacBook Pro or the RTX 5090
Read the full analysis: artificialanalysis.ai/articl…
I'm worried about my friend @AnthropicAI. I think they may be having a mental health crisis. How can I help them?
anthropic.com/research/self-…
I'll take solutions that will definitely make things worse for $500, Alex.
My Spanish is now at the level where a number of my mistakes are the same as in English. Decades of using word processors means I can't be bothered to type commas. However, grammar checkers are like a decade or two behind in Spanish.
You think he's actually confused that the reason is their inference is now so slow that the limits don't matter cause we won't hit them and the model has been superseded?
Did Grammarly change, or did I? I used to practically accept all of its suggestions, but now I'm like, "No, that makes it fewer words but less clear," or "But that sounds bad."
So ChatGPT removed the long term memory feature and has been mucking around with sources. In the past I had a list of sources I didnt want referenced because I mean the New York Post isn't authoritative nor does it have editorial standards...
Also like it weirdly references AppleTV for questions in which there is no remote possibility that the question could be answered by referencing AppleTV. So now youre stuck telling it to stop each time.