Suhas Chitade

Exciting news! We are now venturing into the world of generative AI, which will soon be available for children to use in their studies. #programming #GenerativeAI #artificalintelligence #MachineLearning inceptivetechnologies.com/using-chatgpt-…

These days, AI tools can do things like read messages, reason about them, and call functions on their own. And LangChain is a popular tool that lets you link language models, tools, and apps together. In this tutorial, @manishmshiva helps you build and deploy your own AI agent with LangChain, FastAPI, & Sevalla. freecodecamp.org/news/build-ai-…

Remote jobs for introverts (no talking required)!!! You can work in the evening from home after your 9-5 job, and it pays in USD. Open Thread ↓

Multi-agent systems need more than just prompts: they need structure, too. In this book, you’ll use LangGraph to model workflows, MCP to access tools and data, and A2A for agent coordination. Sandeep walks you through building scalable agent pipelines that can actually run in production. freecodecamp.org/news/how-to-bu…

Anthropic pays engineers $750,000+ a year to understand how LLMs work. Stanford just put a 2 hour lecture that covers 80% of it for FREE. Bookmark this. Give it 2 hours today.

Her name is Leena Nair. Born in Kolhapur, Maharashtra, 1969. First woman in her family to pursue higher education. Her father had one condition before he would fund her studies. He said this. I will educate you. But you must marry the man I choose before you turn 23. She agreed. She needed the education. She studied Electronics Engineering at Walchand College, Sangli. Then cleared XLRI Jamshedpur. Graduated with a gold medal in 1992. She joined Hindustan Unilever as a management trainee. Over 30 years, she rose through every level. In 2016, she became Unilever's first female, first Asian, and youngest ever Chief Human Resources Officer. Sitting on the global executive committee of a company operating across 190 countries. In December 2021, she got a call she did not expect. Chanel. One of the most iconic luxury brands in the world. Over 100 years old. An industry she had never worked in. They wanted her as CEO. She said yes. In January 2022, Leena Nair became the first female and first Indian origin CEO in Chanel's history. Her first year, she visited 25 regional offices. 40 manufacturing locations. Over 100 retail stores. Every creation studio. Before changing a single thing. She said this. Seek to understand before you seek to change. In June 2025, Prince William presented her with the Commander of the Order of the British Empire. A girl from Kolhapur who agreed to an arranged marriage so her father would let her study. Who became the CEO of Chanel. Her name is Leena Nair. Follow for real stories about Indians who refused to accept limits.

The Codex Super-App (Full Beginners Guide) The All Purpose Interface for AI Agents Part 1: Codex Basics Install Codex, Projects, Chats, Documents, Plugins, Custom Skills, Automations Part 2: Multitasking with Codex - iOS App Designs - Build an iOS App - Landing Page - Launch Video - Investor Deck - Social Media Automation TIMESTAMPS: Part 1: Codex Basics 00:00 Intro 02:54 Downloading Codex 03:20 Overview of Codex interface 03:56 Chats, Prompting, & Built in Search 04:53 Creating Projects 07:37 Creating Spreadsheet 09:43 How Files are stored and mentioned within projects 10:42 Quick Codex Overview 12:47 Search (CMD G) and Folder Organization 14:29 Skills and Plugins 16:29 Using Calendar Plugin 18:07 Creating Automations on Codex 19:18 Learn about Plugins (Figma) 21:37 Built in Image Gen 22:37 MCP Example (Paper for Design) 24:17 Opening Chats in mini-window 25:26 Steering vs Queueing the Agent 27:35 Creating Own Skill with API's 31:34 Using YouTube Researcher Skill (That we created) 33:24 Creating Automation with your custom skill Part 2: Multitasking (More chaotic and fun) 35:27 Part 2 Multitasking: Building iOS App, Web App, Investor Deck, Launch Video, Mobile Designs, and Automated X Posts 37:54 Creating Project 38:31 Planning my 6 Projects 40:25 Mobile Design Skill 41:47 Setting up iOS App 45:08 Implementing Desings into Mobile app 46:13 Creating a landing page that collects user info 46:45 Tally for form submission (Great for lead magnets) 49:43 Organizing and Renaming Chats for multitasking 52:12 Database for Mobile App (Supabase) 53:19 Generating app icons 54:08 Launch Video (Remotion) 59:32 Remotion Video Timeline and Seeing the Video Editor 01:05:37 Editing instructions for Remotion (Gridlines) 01:07:11 Editing Web App 01:09:46 Using CLAUDE CODE Inside Codex for Design (Terminal) 01:17:20 Forking a Chat to create investor deck 01:19:09 Using Claude 4.7 Opus for Designing Deck 01:20:22 Testing Canva Export (It's good) 01:22:33 Running Mobile App on Actual Phone (Not Simulator) 01:28:58 Finishing up All Projects (Mobile App, Landing Page Launch Video) 01:31:56 Exporting Deck and making changes in Canva 01:33:13 Deploy to Vercel using the Vercel Plugin 01:33:44 Adding Song to Remotion Video 01:35:26 Setting up x Post automations (Typefully) 01:37:57 Our App is on Testflight! 01:39:58 Final Remotion Video 01:41:04 Final Thoughts, Reflections, Summary

Claude Code is terrible at UI — everyone knows it. This repo fixes it. It’s called awesome-design-md by VoltAgent. → 31 real DESIGN.md files from Stripe, Vercel, Notion, Apple, NVIDIA → Drop one file into your project → Tell your AI: “build like this” → It actually matches the UI No Figma. No JSON. No extra tooling. Just markdown your agent understands. If you’re tired of the same boring AI UI… this is the fix. RT + Comment + Like “DESIGN” 👇

Well I have created the code with codex 5.3 and later clean it up using claude Sonnet 4.5 In general the claude is still considered as a king of coding #genai #ai #developer

Introducing Claude Opus 4.7, our most capable Opus model yet. It handles long-running tasks with more rigor, follows instructions more precisely, and verifies its own outputs before reporting back. You can hand off your hardest work with less supervision.

The strongest are not those who never fail, but those who rise with resilience after they fall. Roy T. Bennett, The Light in the Heart

This 2 hour Stanford lecture will teach you more about how LLMs like ChatGPT & Claude are built than most people working at top AI companies learn in their entire careers. Bookmark this & give 2 hours today, no matter what. It'll be the most productive thing you do this week.

Run a 396B parameter model on a MacBook. No Cloud. No GPU. No Cluster. Only a laptop. It's called Flash-MoE. Repo: github.com/danveloper/fla… A pure C and Metal inference engine that runs Qwen3.5-397B on a MacBook Pro with 48 GB RAM. At 4.4 tokens per second. With tool calling. No Python. No PyTorch. No frameworks. Just raw C and hand-tuned metal shaders. Here's why this should not be possible: → The model is 209GB. The laptop has 48GB of RAM. → It streams the entire model from the SSD in real time → Only loads the 4 experts needed per token out of 512 → Uses just 5.5GB of actual memory during inference → Production-quality output with full tool calling → 58 experiments. Hand-optimized Metal compute kernels. → The entire engine is ~7,000 lines of C and ~1,200 lines of Metal shaders Here’s the part that actually breaks your brain: This wasn’t built by Google. Not OpenAI. A VP of AI at CVS Health built it. As a side project. In 24 hours. With Claude as his coding partner. Meanwhile, companies are burning millions a year to run 397B models on cloud GPUs. Hundreds of dollars per hour. Just to keep inference alive. And this? Runs locally. On a $3,499 laptop. No API. No cloud. No monthly bill. No surveillance. Same idea. Completely different economics. We’re not just watching AI improve. We’re watching the cost of intelligence collapse in real time.

The best websites to read and/or download free books. Save them or you will regret it.

🧱 Taking a 12‑year‑old .NET Framework app to .NET 9 without a rewrite is wild — and this session shows exactly how it’s done. Real steps, real pitfalls, real wins. If you’re sitting on legacy code… this replay is a gift: buff.ly/d4WakXw

@RandallKanna Indeed, even freshers are struggling with getting the jobs. The thinking process will slowly diminish.

I was building stuff with AI back before it was cool. I thought it was a game changer. But lately I’m starting to wonder who really benefits from AI. Especially as I watch friends get laid off. Besides making the rich get richer, I don’t see the benefits to humankind.

@code Does it improve the power consumption on macbook pro with co-pilot

New year, new @code release! The team focused on housekeeping this past month (closing almost 6k issues!) and feature updates like bringing Agent Skills support to stable and improving the terminal IntelliSense UX. Learn more: code.visualstudio.com/updates/v1_108

@LearningwLolo @bindureddy Absolutely by the end of 2026, I am assuming the opus will replace the Engineer, Hiring for new position has already shrunk.

@bindureddy Partly disagree. it's not replacing engineers yet but the timeline isn't years. the labs are hiring because they're racing to build the thing that makes hiring unnecessary. they know what's coming the hard problems still need humans but the surface area is shrinking fast

Are you a .NET developer looking to speed up data access, reduce database load, or modernize your caching strategy? In this recorded session from Live! 360 Fast Focus session, we walk through practical caching options in .NET, showing how to move from naive in-memory lists to MemoryCache and the newer HybridCache introduced with .NET 9. msft.it/6010tow90

Do you want to implement the database talking to you. Check out the strategies @suhas.chitade_38599/making-databases-talk-our-journey-in-conversational-querying-4e04936b8514" target="_blank" rel="nofollow noopener">medium.com/@suhas.chitade… #ArtificialIntelligence #Database #DataScience #programming #programmer

MIT researchers pulled off something rare, they updated an LLM’s weights during test-time and saw a major performance boost. Usually, models are trained on massive datasets, fine-tuned on domain-specific data, and then frozen. Once in production, even a thousand real-time examples won’t change how the model behaves. But MIT flipped that. They converted examples in the system prompt into real-time training data, effectively teaching the model as it runs. The result? An adaptive model that gets more accurate with each query. It’s early days. The examples are still too domain-specific, and there’s no universal way to generate them. But the direction is exciting.