Ferdl S

1.1K posts

Ferdl S banner
Ferdl S

Ferdl S

@BavariaR

Supercar lover, AI and Telecommunication Expert, spending lots of time with Topics like Stocks, Finance, IT and Business consulting . Hobby: Paramedic@Red Cross

Haarbach Katılım Ağustos 2009
339 Takip Edilen73 Takipçiler
Ferdl S
Ferdl S@BavariaR·
Is it possible to use the Codex plugin even if you're not using Claude Code for coding, but for business tasks? Can you get a second opinion that way, or — like what's become pretty common now — when you run out of tokens mid-month on your Claude Code account, can you then continue working with Codex, which would then follow the Claude Code rules and approaches within the Claude Code environment? Or is it really only meant for coding?
English
0
0
0
29
Kai
Kai@hqmank·
If you're using Claude Code, this is worth knowing. Instead of worrying about whether Opus 4.6 or GPT 5.4 is better, it's more useful to combine them in the same workflow. OpenAI shipped an official Claude Code plugin called codex-plugin-cc. You can now call Codex directly from inside Claude Code. Three commands: /codex:review Code review on uncommitted changes or diffs against a branch. Read-only. /codex:adversarial-review Challenges your design decisions, not just syntax. "Why this caching strategy?" "Race condition here?" Append free-form text to steer the review. /codex:rescue Hands the task to Codex when Claude gets stuck. Supports --resume to continue from the last run. Adversarial review is the killer feature. Especially before shipping auth changes, infra scripts, or anything involving data loss. There's also a review gate: Codex auto-reviews every time Claude finishes and blocks completion if issues are found. Claude writes, Codex reviews. github.com/openai/codex-p…
Vaibhav (VB) Srivastav@reach_vb

x.com/i/article/2038…

English
75
174
2.2K
490.2K
Ferdl S
Ferdl S@BavariaR·
What really gets me is that I actually had more trust in Anthropic than in other LLM providers — that they think transparently, honestly, with a focus on safety, and that they act ethically. And what I'm seeing right now with this whole situation, where they're burning through tokens faster to deal with their own problem — namely that they don't have enough resources to handle the influx of new users — while screwing over existing users and then lying to them on top of it... That has nothing to do with ethics, it's not honest, and I feel completely played.
English
0
0
0
6
Dan ⚡️
Dan ⚡️@d4m1n·
Anthropic says hitting rate limits so fast in Claude Code is your fault. do they think we all just subscribed this week?!
Dan ⚡️ tweet media
English
122
32
905
33.9K
Ferdl S
Ferdl S@BavariaR·
@bcherny Yeah, just testing it but I feel that it's really still better. The flickering is obviously improved but I have difficulties copying and pasting text or pictures into the terminal window, especially under VS Code
English
0
0
0
6
Boris Cherny
Boris Cherny@bcherny·
Today we're excited to announce NO_FLICKER mode for Claude Code in the terminal It uses an experimental new renderer that we're excited about. The renderer is early and has tradeoffs, but already we've found that most internal users prefer it over the old renderer. It also supports mouse events (yes, in a terminal). Try it: CLAUDE_CODE_NO_FLICKER=1 claude
Curt Tigges@CurtTigges

@bcherny @UltraLinx please at least fix the uncontrollable scrolling/flickering before the next 3000 features

English
630
681
10K
2.4M
Ferdl S
Ferdl S@BavariaR·
Yeah, definitely looking for something like a dashboard which can overlook all my Anthropic work, what I do in Claude Code, what I do in Claude CoWork and in chat. It can also manage several sessions in parallel and be able to understand the overall context of what I'm doing there... I really start to lose the oversight.
English
0
0
0
30
Ferdl S
Ferdl S@BavariaR·
@testingcatalog Still waiting for getting the Claude Code desktop update on Windows, which includes computer. Actually the last update I got today did not yet include computer on Windows so it should come as an Easter present
English
0
0
0
138
International Cyber Digest
International Cyber Digest@IntCyberDigest·
🚨‼️ BREAKING: Anthropic has decided to open source their entire codebase and is rebranding their AI to OpenClaude. Anthropic CEO Dario Amodei said: "Yesterday was no slip-up. If we disappear just like OpenAI is vanishing right now, our code can live on through the community."
International Cyber Digest tweet mediaInternational Cyber Digest tweet media
English
983
2K
17.4K
2.5M
Ferdl S
Ferdl S@BavariaR·
Claude / Perplexity Computer: Both are AI-powered computer use / agent tools, but they take pretty different approaches: Claude Computer (by Anthropic) - Focuses on actually controlling your desktop — clicking, typing, navigating apps, running tasks autonomously - Built around Claude's reasoning capabilities to handle complex, multi-step workflows - More of a general-purpose agent that can interact with basically anything on your screen Perplexity Computer (by Perplexity AI) - More search and research-oriented at its core - Integrates web search deeply into its workflow - Less focused on full desktop control, more on finding and synthesizing information So while both fall under the broad "AI agent" umbrella, Claude Computer leans more toward task automation and desktop control, while Perplexity Computer leans more toward research and information retrieval. Similar concept on the surface, but different strengths in practice!
English
0
0
1
61
Poonam Soni
Poonam Soni@CodeByPoonam·
🚨Claude just broke the AI industry. Again. This time it's sitting at your desk, running your apps, doing your work. Text it from your phone. Come back to finished work. That's literally it. Here's what it can do (🔖 Save this cheatsheet): → Open apps, find files, export, attach, send → Fill spreadsheets from browser tabs → Schedule emails and morning briefings → Submit PRs and run tests → Organize your Downloads folder → Update your calendar from Slack messages
Poonam Soni tweet media
English
13
11
54
3.9K
Ferdl S
Ferdl S@BavariaR·
That's probably also why there are no resources left for us — we end up getting throttled during peak hours and our credits burn through faster, because Anthropic is consuming all their own resources to develop their internal tools. And then we, the paying customers, get punished for it. Great, thanks for that.
English
0
0
0
38
CG
CG@cgtwts·
Anthropic is using its own AI tool Claude Code, like crazy. In just 52 days, they pushed out 50+ new updates One employee used it so much in a month, it cost $150,000. Around 80% of the team uses it every day, and the heaviest users are spending hundreds of thousands on it.
CG tweet media
CG@cgtwts

Anthropic CEO: “ I have engineers within anthropic who don’t write any code, they just let Claude write the code and they edit it and look it over” “At anthropic writing code means designing the next version of Claude it self, so we essentially have Claude designing the next version of Claude itself, not completely but most of it”. In the last 52 days, the Claude team dropped 50+ major feature launches. This is literally INSANE.

English
46
15
167
29.7K
Ferdl S
Ferdl S@BavariaR·
I can only second this experience. I had disagreements with the billing of the API interface where money disappeared without usage being reported. I have tried two times to settle this. I escalated it two times to a human, and nothing really happened, so I was just happy that it was about a small amount.
English
0
0
3
1.8K
adriansio
adriansio@adriansio·
Boris, I was probably your biggest fan. Two Max accounts, $400/mo. Yesterday your billing system charged my card 75 times in 12 minutes (~$615). I reported the bug. Anthropic response? Suspended both my accounts. Your own billing system malfunctions, I report it, and I get punished for it. No human has responded to my support emails or appeal forms. Just an AI chatbot trying to close my ticket. I'm still a fan of the product. But your billing and support systems are broken and your paying customers deserve better than automated bans and radio silence.
English
17
5
409
71.7K
Boris Cherny
Boris Cherny@bcherny·
A weird part of working at Anthropic: getting a few of these each day
Boris Cherny tweet media
English
520
121
8.1K
1M
Ferdl S
Ferdl S@BavariaR·
Ich freue mich darüber dass Infomaniak ein multilinguales Unternehmen ist. Ich möchte nur anmerken, dass alle Hilfstexte ziemlich schlecht auf Deutsch übersetzt sind. Gott sei dank verstehe ich Französisch und suche immer die französischen raus, um sicher zu sein, dass ich auch verstehe, was gemeint ist. Die deutsche Übersetzung sollte mal EURIA gegeben werden. Vielleicht schafft die es, eine bessere Übersetzung zu erstellen!
Deutsch
1
0
0
19
Infomaniak DE
Infomaniak DE@infomaniak_de·
Infomaniak kennt keinen Röstigraben! Seit mehr als 9 Jahren sind wir mit einem Büro in der Deutschschweiz präsent – ganz nah an unseren Kunden. Wer Infomaniak kennt, weiss: Was uns stark macht, sind nicht nur unsere Infrastrukturen und Dienste, sondern auch die Atmosphäre in unseren Büros und die Unternehmenskultur, die wir seit über 30 Jahren pflegen. All das mit einem Team zu teilen, das mehrere Hundert Kilometer entfernt arbeitet, ist nicht immer einfach. Diese Woche war unser Zürcher Team zu Gast in Genf: ✅ Besichtigung unseres Rechenzentrums ✅ Schulungen mit unseren lokalen Teams ✅ direkter Austausch mit unseren Ingenieuren und Entwicklern Schöne Momente, die Verbindungen stärken, Synergien schaffen und die Zusammenarbeit zum Wohl unserer Kunden verbessern. Werden Sie Teil unseres Teams: infomaniak.com/gtl/about.jobs
Infomaniak DE tweet mediaInfomaniak DE tweet mediaInfomaniak DE tweet media
Deutsch
1
2
3
96
Ferdl S
Ferdl S@BavariaR·
The problem that still exists with Claude-Corework and Scheduled Tasks is that it doesn't reliably teach the task permissions. Even when you follow the instructions exactly and run the task manually to grant permissions like deleting files, these permissions aren't actually assigned to the task itself but to the specific instance that's currently running within the task. The permission isn't carried over to subsequent task runs. In my opinion, that's a massive bug.
English
0
0
0
4
Claude
Claude@claudeai·
New in Cowork: scheduled tasks. Claude can now complete recurring tasks at specific times automatically: a morning brief, weekly spreadsheet updates, Friday team presentations.
English
989
1.7K
22.4K
8.2M
Ferdl S
Ferdl S@BavariaR·
Wenn man die Rente für Gutverdiener deckelt, sollte man auch die Option haben, dieses Investment in eine Rente zu überdenken. Denn dann muss ich so oder so selbst für meine zusätzliche Rente sorgen – und bin nicht mehr bereit, in das System einzuzahlen. Ich kaufe ja auch keine Schrott-Aktien, wenn ich nichts davon habe. Wenn die Rendite der gesetzlichen Rente so schlecht wird, muss ich auch die Möglichkeit haben, darauf zu verzichten.
Deutsch
0
0
0
10
sparbuchfeinde
sparbuchfeinde@sparbuchfeinde·
So kann die SPD die an die AfD verlorenen Arbeiterstimmen sicher zurückgewinnen. Industriemeister mit 45 Beitragsjahren soll die Rente gedeckelt bekommen. Was für ein Wahnsinn!
sparbuchfeinde tweet media
Deutsch
295
497
3.1K
56.5K
Ferdl S
Ferdl S@BavariaR·
@sparbuchfeinde Was heißt hier deckeln? Sogar wenn man gut verdient ist, ist die Rente ein Witz in Deutschland. Schaut mal rüber nach Luxemburg oder nach Österreich. In Österreich sind die Renten mindestens 50 % höher, in Luxemburg sind es über 100 %
Deutsch
0
0
1
10
Ferdl S
Ferdl S@BavariaR·
@mario_lochner Es kommt mir immer mehr so vor, als würde Trump gezielt nach günstigen Einstiegsmöglichkeiten an der Börse suchen. Er haut eine Drohung raus, kauft munter und günstig ein, macht einen schönen Reibach – und spielt dabei mit der Sicherheit der Weltpolitik.
Deutsch
0
0
1
49
Mario Lochner
Mario Lochner@mario_lochner·
Iranische Medien berichten, dass es keinen direkten oder indirekten Kontakt mit Trump gab, und behaupten, er habe einen Rückzieher gemacht Es bleibt spannend...
Deutsch
27
8
254
18.3K
Ferdl S
Ferdl S@BavariaR·
I'm still struggling to understand exactly how both complement each other cleanly. When I integrate a project from Claude Chat into Claude Cowork, it's still not clear to me whether that's a migration or just a copy of the project that exists in Chat. The feature that would be most important to me in Projects is being able to maintain context across multiple chat sessions in Claude Cowork. Whether it makes sense to adopt the file repository from Projects into Cowork - I still have my doubts about that, because I could just put the same files in a folder on my computer and have them locally. I honestly haven't really understood what the advantages are there. All the explanations I've tried to find so far haven't actually answered my question.
English
0
0
1
27
Wes Roth
Wes Roth@WesRoth·
Anthropic rolled out the "Projects" feature for its local Claude Cowork desktop environment. Users can now organize their tasks, files, and custom instructions into focused, project-specific hubs, eliminating the need to constantly re-upload context for ongoing workflows.
Wes Roth tweet media
Claude@claudeai

Projects are now available in Cowork. Keep your tasks and context in one place, focused on one area of work. Files and instructions stay on your computer. Import existing projects in one click, or start fresh.

English
11
13
152
12.3K
Ferdl S
Ferdl S@BavariaR·
Once we have a finished prototype or even a product in Germany, we won't find any buyers here again. The government won't have money to bring the project into reality anywhere. The states will say they can't afford a project like this without cross-financing from the federal government to install it somewhere. We've already been through this exact same mess with a high-speed rail system. And it's going to play out the same way as last time. The Chinese will buy three of these projects, implement two of them, completely disassemble one, copy the technology, and happily continue developing it themselves. We'll have invented everything again and get nothing out of it because we don't implement it in reality and because nobody in Germany wants to spend money on it.
English
0
0
1
24
Ferdl S
Ferdl S@BavariaR·
@mario_lochner Schade dass die FDP es nicht schafft wieder hoch zu kommen. So eine Partei der Mitte wäre sehr hilfreich um die Balance zwischen rechts und links besser zu managen
Deutsch
0
0
1
133
Mario Lochner
Mario Lochner@mario_lochner·
Gute Nachrichten: SPD bekommt eine saftige Quittung, Grüne verlieren und die Linken schaffen es nicht in den Landtag #rheinlandpfalz
Mario Lochner tweet media
Deutsch
109
35
1.3K
33.5K
Freya Lawson
Freya Lawson@Freyabuilds·
🚨BREAKING: You can now run Claude Code for FREE. No API costs. No rate limits. 100% local on your machine. Here's how to run Claude Code locally (100% free & fully private):
Freya Lawson tweet media
English
4
22
99
16.6K
Abhishek
Abhishek@HeyAbhishek·
Most people don't know Claude has 3 different products. They keep using Chat for everything and wonder why it's slow. Here's when to use each one:
Abhishek tweet media
English
4
12
88
14.7K
Ferdl S
Ferdl S@BavariaR·
@sukh_saroy Why are you teasing with Claude if you explain how to install open source LLMs which are not to be compared with Claude?
English
0
0
1
69
Sukh Sroay
Sukh Sroay@sukh_saroy·
🚨BREAKING: You can now run Claude Code for FREE. No API costs. No rate limits. 100% local on your machine. Here's how to run Claude Code locally (100% free & fully private):
Sukh Sroay tweet media
English
8
21
101
16K