CX Data Scientist 📊 by day ☀️ Indie hacker by night 🌙 Web & App Dev. All things AI 🤖 Also 🇯🇵 👨👩👧👦☕🍵🚀
Wiesbaden, Germany
Joined September 2010
- Tweets7.7K
- Following597
- Followers2.2K
- Likes15.3K
Nice black mirror-like idea but why use a cheaply generated AI image for this? The description sounds like cyber punk dogs and the pic looks like paw patrol for grown ups lol 😂
Palmer Luckey unveils “Warhound,” a $200 million startup developing genetically enhanced dogs for the US military in partnership with Colossal Biosciences and Anduril Industries.
Colossal will provide genetic engineering and cloning expertise, while Anduril will equip the dogs with advanced cybernetic technology and battlefield awareness systems.
Follow: @AFpost
Is it possible that all three memes are correlated and distance from actual work is causing all three?!😅
Downgraded my chatGPT 5x Pro back to Plus. Its not that I don't like it, it helped me to work on two game websites and got me pretty far, they're basically live, but the huge drawbacks were:
(1) I liked working with Astra, but even with medium reasoning token-burn was bad and
(2) more severe, even Astra couldn't handle frontend well. I always needed to use Opus to redo all frontend things Astra built. Maybe a skill issue, but I don't want to use elaborate prompts or skills or what ever, I just tell Opus to make it look nice, and the first shot is usually pretty good, then I do two three more rounds and the design as well as the UX are pretty much perfect (for me).
What I really liked about Astra, and what I will definitely miss, was its computer use. You could just tell it to set things up: healthchecks .io, Cloudflare, Discord, you call it, it just opened safari and did its thing. Sometimes asking for permissions, but in general pretty smooth.
Let's see what kind of updates Pro users receive in the next weeks, maybe they're gonna win me back. For now I'm thinking about getting Claude 5x Max.
Looks like I wasn’t so wrong.
I’m not a fan boy of any particular lab, but I highly doubt this is a meaningful benchmark. We would need to compare how much dollars of a subscription different tasks cost. And a result might be, that Antrhopic is much cheaper. But considering OpenAIs token effiency, and also that OpenAI token ≠ Anthropic token this is not clear yet, and this benchmark might be misleading.
Well idk why but until Opus 5 there was a fast mode in the desktop app, maybe its still there in the CLI, wonder why its not available for 5.5.
Looks like the X-Wings arrive just in time!
🚨 BREAKING: Google have announced Gemini 4 Argon, their new flagship model currently being tested in partnership with the US Government before a wider launch
It's the new best model in the world. What a turnaround
blog.google/innovation-and-a…
@AnthropicAI I'd gladly participate in your interview, but you forgot one important point to add to your invitation ↓
