احصائيات اقرأ-تِك لشهر فبراير 2025 🥳 من فترة للتانية بحب أنزل احصائيات اقرأ-تِك عشان أذكر نفسي والجميع بأهمية اللي بيعملوه وإن الاستمرارية حقيقي على المدى الطويل مهما كانت الخطوات صغيرة بتكون جميلة جدًا .. وكل ده هو فضل من ربنا وكرم منه فاللهم لك الحمد ❤️ النشرة الأسبوعية 📮

أنا معرفش إزاي فضلت ٤ ساعات في تسجيل غير الوقت قبلها في التحضير بس بصراحة كانت حاجة عظمة، أما تجيب موديل كحوول زي qwen 2.5 1.5b و تطلب منه تاسكات و ميعرفش يعملها و بعد كده تجيب gpt4-o موديل و تخرج منه شوية داتا نضيفة و تعلم كوين عليها فيبقى قادر يعمل التاسك كويس، فأنا فتحت بقي 😃

اللهم لك الحمد.. يسعدني أن أعلن عن انتهائي من رفع و تسجيل كافة دروس سلسلة react js بواقع ٢٨ ساعة تعليمية، و التي عملت عليها خلال ال ٣ شهور الأخيرة 🙌🏻😄 كما يسعدني أن أعلن عن توفيرها بشكل مجاني لكل مبتدئ في البرمجة ضمن مسار تعليمي متكامل لل frontend🎉😄 كيف؟ إليك التفاصيل.. 👇🏻

مساء الخير يا مدينة البط وحشنى جدا عمتا عم دهب هو السبب المش بيخلينى انزل اعداد كتير عشان الbusiness needs المهم عشان النهارده الخميس هنزل عدد من اعداد سوبر ميكى سنة 1999 من دار الهلال العدد فى الثريد

Best-of-N Jailbreaking is a black-bok algorithm with an attack success rate of 89% on GPT-4o and 78% on Claude 3.5 Sonnet. The jailbreaking technique combines augmentations such as random shuffling or capitalization. It can also be extended to jailbreak vision and audio

ChatGPT is just an HTML web page with a simple input box. I don't see what's a big deal - any decent web developer could probably replicate it in under an hour.

Composition of Experts Proposes Composition of Experts, an efficient modular compound AI system leveraging multiple expert LLMs. A router helps select the right expert for a given input, enabling efficient resource utilization and improving performance. While the general

This paper comes back to show again, that LLMs are just pattern-matching machines, not step-by-step reasoners when solving math problems. Reveals that LLMs don't actually perform step-by-step reasoning when generating direct answers without showing their work. Through probing

Everyone’s taking about Agents but few are demoing. Watch me build an AI Agent in 5min that has already helped us make a killer hire at @glideapps 💪

Follow to me and give me more motivation to update my products.

I'm building the same app using all popular AI IDEs. Their progress is insane > Replit got way better at UI > Cursor shipped coding agents > Windsurf entered the game and made a nice > V0 can do full-stack apps > Bolt is turning noncoders into coders Mega thread on AI IDEs🧵:

How to create a full frontend from one design: 1. Screenshot the design 2. Generate prompts in CopyCoder 3. Paste prompts into Cursor and keep typing "yes" until the build is done → Full frontend app with all the pages created

RAG data chunking in a few lines of Python Code. Installs in seconds and just works. 100% opensource.

Another great practical resource on Postgres, straight up from their official wiki: Don't do this contains a bunch of things you should *not* be doing

After the release of Qwen2.5, we heard the community’s demand for processing longer contexts. https://t.co/zu2mOB9CQm Today, we are proud to introduce the new Qwen2.5-Turbo version, which features: 📚 Longer Context Support: We have extended the model’s context length from

I love doing these kind of recommendations! Stop using EasyOCR ( https://t.co/ex7OP3g4xL), it's easy to use yes but a bit outdated (4/5 years old). Try MGP-STR instead, a ViT based model available on @huggingface for running OCR at the edge! https://t.co/zgVTdIA99V

Everyone thinks this is an exaggeration but there are so many software engineers, not just at FAANG, who I know personally who literally make ~2 code changes a month, few emails, few meetings, remote work, < 5 hours/ week, for ~$200-300k. Here are some of those companies:

AWS just released a new Multi-Agent AI framework It lets you manage multiple AI agents, dynamically route LLM queries, maintain context across AI Agents and can be deployed locally on your computer. 100% opensource.

Announcing llama-ocr – a free + open source OCR tool! It takes documents (images for now) & outputs markdown, and does really well for complex receipts, PDFs with tables/charts, ect... Powered by Llama 3.2 vision on @togethercompute & available on npm today!

5 years ago, I needed 30 people & $3M in funding to build just one startup, Today, I run 11 bootstrapped SaaS projects with just a handful of people. These AI agents & tools 10x busy founders in marketing, coding, legal, operations, design, sales, anything 🧵 :