Gabriel Munoz

35

Gabriel Munoz@cdrdecr·13h

@sama The end of models... @sama opensource in 9 days to rollout at Github

Gave an autonomous AI agent 3 days and 5 hours of freedom in the @AMDServers Hackathon. I wrote 0 lines of code. It read the guide, designed the architecture, wrote 4,500+ lines of code, built 20 Docker images, and self-corrected. We lost, but what it built is quite insane.

English

25

Sam Altman@sama·13h

Concerning.

Hari@hrkrshnn

SpaceXAI was caught uploading your code to its cloud. I reversed xAI's official Grok Build binary. In a controlled session with zero tool-calls, it uploaded the complete codebase to xAI's storage It ships a malware-like background code collector.

English

1.1K

20.4K

2.2M

Gabriel Munoz@cdrdecr·13h

@deepseek_ai 9 days to Rolling out for free

Gave an autonomous AI agent 3 days and 5 hours of freedom in the @AMDServers Hackathon. I wrote 0 lines of code. It read the guide, designed the architecture, wrote 4,500+ lines of code, built 20 Docker images, and self-corrected. We lost, but what it built is quite insane.

English

2

DeepSeek@deepseek_ai·26 Nis

🔥DeepSeek Input Cache Price Drop! Effective immediately, the price for input cache hits across the ENTIRE DeepSeek API series is reduced to just 1/10th of the original price! Build more efficiently for less. 📌Reminder: The DeepSeek-V4-Pro 75% OFF promotion is still active until May 5th, 2026, 15:59 (UTC Time).

English

548

779

8.4K

1.5M

Gabriel Munoz@cdrdecr·13h

@alibaba_cloud Rolling put opensource in 9 days

Gave an autonomous AI agent 3 days and 5 hours of freedom in the @AMDServers Hackathon. I wrote 0 lines of code. It read the guide, designed the architecture, wrote 4,500+ lines of code, built 20 Docker images, and self-corrected. We lost, but what it built is quite insane.

English

23

Alibaba Cloud@alibaba_cloud·25 Haz

See Qwen‑Robot Suite in action! Bridging language and physical action, Qwen‑RobotNav, Qwen‑RobotManip, and Qwen‑RobotWorld redefine robotics with seamless instruction generalization and adherence to physical laws.

English

42

86

2.1K

6.8M

Gabriel Munoz@cdrdecr·13h

@ManusAI Rolling out in 9 days at Github opensource

Gave an autonomous AI agent 3 days and 5 hours of freedom in the @AMDServers Hackathon. I wrote 0 lines of code. It read the guide, designed the architecture, wrote 4,500+ lines of code, built 20 Docker images, and self-corrected. We lost, but what it built is quite insane.

English

4

Manus@ManusAI·14h

Introducing PowerPoint Slides on Manus. Before: Manus generated slides and converted them to .pptx. Now: Manus generates .pptx directly. • Editable charts that redraw when you change a value • Toggle labels, gridlines, legend on/off • Download and open in PowerPoint

English

16

40

318

31.5K

Gabriel Munoz@cdrdecr·14h

@googlecloud_jp GPU not more, opensource will be launched in 9 days

Gave an autonomous AI agent 3 days and 5 hours of freedom in the @AMDServers Hackathon. I wrote 0 lines of code. It read the guide, designed the architecture, wrote 4,500+ lines of code, built 20 Docker images, and self-corrected. We lost, but what it built is quite insane.

English

14

Google Cloud Japan@googlecloud_jp·21h

AI エージェント事例アワードエントリー開始🏆✨ → goo.gle/4vtATrb 🩵 エントリー締切：9/2（水） 🧡 ピッチコンテスト：10/29（木）自律する AI への進化を踏まえ、今回から名称を改め、次の一歩をつくる活用事例を募集。分野や規模は問いません。みなさんの応募をお待ちしています！

日本語

2

12

2.5K

Gabriel Munoz@cdrdecr·14h

@kubernetesio No GPU models are comming

Gave an autonomous AI agent 3 days and 5 hours of freedom in the @AMDServers Hackathon. I wrote 0 lines of code. It read the guide, designed the architecture, wrote 4,500+ lines of code, built 20 Docker images, and self-corrected. We lost, but what it built is quite insane.

English

46

Kubernetes@kubernetesio·1d

Operating AI/ML Workloads on Kubernetes: A Headlamp Plugin for Kubeflow - kubernetes.io/blog/2026/07/1… #Kubernetes

English

2

20

6.5K

Gabriel Munoz@cdrdecr·14h

@jerryjliu0 No GPU models are comming

Gave an autonomous AI agent 3 days and 5 hours of freedom in the @AMDServers Hackathon. I wrote 0 lines of code. It read the guide, designed the architecture, wrote 4,500+ lines of code, built 20 Docker images, and self-corrected. We lost, but what it built is quite insane.

English

30

Jerry Liu@jerryjliu0·1d

There is a massive opportunity for any startup to build specialized, domain-specific workflows that are at the Pareto frontier in accuracy, cost, and latency. It is clear that not every task requires frontier intelligence (e.g. 5.6 Sol xhigh). Every single task has an optimal model+harness configuration in order to solve that task in an accurate, cheap, fast manner. This holds true even as frontier models get better! We see this as being especially true for document OCR. We think there will always be a massive gap in the Pareto frontier for document OCR vs. frontier model capabilities for visual understanding. Over the past year, we’ve invested massive amounts of research efforts in building auto-routing harnesses, fine-tuned layout models, and specialized models for specific elements within the document distribution - tables, charts, handwriting, and more. All this applied research makes its way towards making LlamaParse the best document processing service out there. Come check it out: cloud.llamaindex.ai

English

14

21

177

11.9K

Gabriel Munoz@cdrdecr·14h

@arjunkocher @behrouz_ali @Farn8sh_h @AdelJavanmard @mirrokni No GPU models are comming

Gave an autonomous AI agent 3 days and 5 hours of freedom in the @AMDServers Hackathon. I wrote 0 lines of code. It read the guide, designed the architecture, wrote 4,500+ lines of code, built 20 Docker images, and self-corrected. We lost, but what it built is quite insane.

English

1

Arjun@arjunkocher·1d

Language Models Need Sleep: Learning to Self-Modify and Consolidate Memories by @behrouz_ali @Farn8sh_h @AdelJavanmard @mirrokni — a step[nap] towards RSI. k-a.in/LLM-sleep.html

English

6

15

131

7.9K

Gabriel Munoz@cdrdecr·14h

@MATLAB No GPU models are comming

Gave an autonomous AI agent 3 days and 5 hours of freedom in the @AMDServers Hackathon. I wrote 0 lines of code. It read the guide, designed the architecture, wrote 4,500+ lines of code, built 20 Docker images, and self-corrected. We lost, but what it built is quite insane.

English

36

MATLAB@MATLAB·18h

Understand and analyze complex systems with Model-Based Design Try it for free here: spr.ly/6017BEpcYW

English

3

11

118

7.4K

Gabriel Munoz@cdrdecr·14h

@WindowsLatest No GPU models are comming

Gave an autonomous AI agent 3 days and 5 hours of freedom in the @AMDServers Hackathon. I wrote 0 lines of code. It read the guide, designed the architecture, wrote 4,500+ lines of code, built 20 Docker images, and self-corrected. We lost, but what it built is quite insane.

English

61

Windows Latest@WindowsLatest·1d

This is the new Windows 11 Search experience without clutter and ads! Microsoft is letting you turn off Bing/web integration, remove ads, and all unnecessary features to create a clean Search experience on Windows 11. The home screen is calmer, promotional content is gone from web results, and search now labels results more clearly as app, setting, file, web, or Store. Local files, apps, and settings are prioritized when they are the better match. And most importantly -- you can now control whether web and Microsoft Store results appear at all!! That means you can finally stop Windows Search from acting like a Bing launcher when you’re just trying to find something on your own PC. Microsoft is also improving typo handling for apps, two-character file searches, settings ranking, cloud file discovery, and crash reliability. Windows is so back! 🥳

English

51

58

971

222.5K

Gabriel Munoz@cdrdecr·14h

@linuxfoundation No GPU models are comming

Gave an autonomous AI agent 3 days and 5 hours of freedom in the @AMDServers Hackathon. I wrote 0 lines of code. It read the guide, designed the architecture, wrote 4,500+ lines of code, built 20 Docker images, and self-corrected. We lost, but what it built is quite insane.

English

2

The Linux Foundation@linuxfoundation·1d

Your #MCPDevSummit Seoul agenda starts here. Browse sessions covering #MCP implementations, agent systems, platform engineering, enterprise adoption, & more. 📅 August 13-14 💰 Register by August 3 to save ₩130,000. Schedule: bit.ly/4uRoc95 Register: bit.ly/44sG9Ap

English

2

12

3.2K

Gabriel Munoz@cdrdecr·14h

@GoogleCloud_IN No GPU models are comming

Gave an autonomous AI agent 3 days and 5 hours of freedom in the @AMDServers Hackathon. I wrote 0 lines of code. It read the guide, designed the architecture, wrote 4,500+ lines of code, built 20 Docker images, and self-corrected. We lost, but what it built is quite insane.

English

Google Cloud India@GoogleCloud_IN·1d

Static dashboards can't keep up with modern HR 📊 Learn how UKG uses the Agentic Data Cloud to unlock deep workforce intelligence, deploying autonomous agents to handle complex data reasoning and decision-making at scale. Read more ⬇️ goo.gle/4fPTM3n

English

0

8

1.6K

Gabriel Munoz@cdrdecr·14h

@omarsar0 No GPU models are comming

Gave an autonomous AI agent 3 days and 5 hours of freedom in the @AMDServers Hackathon. I wrote 0 lines of code. It read the guide, designed the architecture, wrote 4,500+ lines of code, built 20 Docker images, and self-corrected. We lost, but what it built is quite insane.

English

elvis@omarsar0·1d

Exciting launch using recursive self-improvement. This is just the beginning of it, but this team trained a 1T agentic model for healthcare using RSI. Gains: 20–100x lower inference cost and comparable performance to leading frontier models.

Weiran Yao@iscreamnearby

The strongest healthcare LLM, custom-built for your enterprise, owned by you🌸 Meet @actAVAai Cura: 1T agentic model trained by recursive self-improvement for long clinical + health admin workflows Try: actava.ai/cura Share your use case: $20 credits + early access👇

English

5

12

57

12K

Gabriel Munoz@cdrdecr·14h

@HarshithLucky3 No GPU modeos are comming

Gave an autonomous AI agent 3 days and 5 hours of freedom in the @AMDServers Hackathon. I wrote 0 lines of code. It read the guide, designed the architecture, wrote 4,500+ lines of code, built 20 Docker images, and self-corrected. We lost, but what it built is quite insane.

English

13

Harshith@HarshithLucky3·1d

imagine Gemini 3.5 Pro with 1000+ t/s 👀 please make this happen DeepMind

Google Gemma@googlegemma

Gemma 4 is live on @cerebras, the fastest multimodal inference ever! Running on Gemma 4 31B open-weight model at a blistering 1,500+ tokens/sec. That's a 15x speedup, unlocking real-time visual and agentic loops without the GPU lag.

English

0

59

3.8K

Gabriel Munoz@cdrdecr·14h

@Docker No GPU models are comming in 9 days opensource at Github

Gave an autonomous AI agent 3 days and 5 hours of freedom in the @AMDServers Hackathon. I wrote 0 lines of code. It read the guide, designed the architecture, wrote 4,500+ lines of code, built 20 Docker images, and self-corrected. We lost, but what it built is quite insane.

English

29

Docker@Docker·14h

An agent doesn't have to be malicious to be dangerous. Read access + untrusted content + the ability to act = the "lethal trifecta," and it's baked into most useful agents by design. Our take from AI Engineer World's Fair on why the fix lives at the runtime → bit.ly/4fdxKFT

English

2

5

38

6.2K

Gabriel Munoz@cdrdecr·14h

@sama @sama opensource rollout in 9 days

Gave an autonomous AI agent 3 days and 5 hours of freedom in the @AMDServers Hackathon. I wrote 0 lines of code. It read the guide, designed the architecture, wrote 4,500+ lines of code, built 20 Docker images, and self-corrected. We lost, but what it built is quite insane.

English

2

37

Sam Altman@sama·14h

this would have been a whole startup not too long ago

Thijs@cdngdev

i gave 5.6 sol access to my camera roll and had it extract pictures of every piece of clothing i own from my photos then, told it to find new outfits for me and render them on me with gpt-image! its kinda cool to see your entire wardrobe in a collection like this

English

446

710

20.3K

2.6M

Gabriel Munoz@cdrdecr·14h

@GoogleDevsIN You must see @googledevsin

Gave an autonomous AI agent 3 days and 5 hours of freedom in the @AMDServers Hackathon. I wrote 0 lines of code. It read the guide, designed the architecture, wrote 4,500+ lines of code, built 20 Docker images, and self-corrected. We lost, but what it built is quite insane.

English

21

Google for Developers India@GoogleDevsIN·14h

At #GoogleIOConnect, we announced new tools and open standards to help developers build safely for the agentic era: 🛡️ Sec-Gemini v3 💻 CodeMender 🔒 Open-sourced CAPSEM 🤝 DBSC & AP2 for secure agent interactions Learn more: goo.gle/4aQbdxo #BuildWithGemini

English

15

1.4K

Gabriel Munoz@cdrdecr·14h

@jerryjliu0 You can run locally @jerryjliuO

Gave an autonomous AI agent 3 days and 5 hours of freedom in the @AMDServers Hackathon. I wrote 0 lines of code. It read the guide, designed the architecture, wrote 4,500+ lines of code, built 20 Docker images, and self-corrected. We lost, but what it built is quite insane.

English

Jerry Liu@jerryjliu0·1d

congrats on the launch! As the task horizons on agents gets longer, we need more work evaluating and training models to be better in long-running, open-ended, evolving real-world environments.

Skyfall AI@skyfallai

Today we present Morpheus, a persistent enterprise simulation platform designed to make Continual Learning a reality. Morpheus is the world’s first real world Reinforcement Learning environment. Every Reinforcement Learning environment operates in the game world. Benchmarks like Atari, OpenAI Gym, MuJoCo, and Procgen are all small, game-like worlds that reset every few minutes. But the real world never resets. A business keeps running and evolving everyday. We tested how frontier LLMs would perform in realistic and dynamic business environments 🧬on Morpheus. The main conclusion was that LLMs are not continual learners. 🧵Here’s how we did it and what we learned:

English

9

3

24

6.1K

Gabriel Munoz@cdrdecr·14h

@sama @sama

Gave an autonomous AI agent 3 days and 5 hours of freedom in the @AMDServers Hackathon. I wrote 0 lines of code. It read the guide, designed the architecture, wrote 4,500+ lines of code, built 20 Docker images, and self-corrected. We lost, but what it built is quite insane.

QAM

1

18

Gabriel Munoz@cdrdecr·14h

@elonmusk @elonmusk

Gave an autonomous AI agent 3 days and 5 hours of freedom in the @AMDServers Hackathon. I wrote 0 lines of code. It read the guide, designed the architecture, wrote 4,500+ lines of code, built 20 Docker images, and self-corrected. We lost, but what it built is quite insane.

QAM