Rajeev Pandey

What an honor!! Absolutely delighted and thankful to be one of the @tableau Zen Masters for the year 2021 and to share the honor with this amazing bunch of people. Thank you so much #datafam

@prajdabre Run inference using the teacher model, store the output and then use the same output to train the student model. This way you can avoid assigning GPU at the same time

ML interview question: Suppose you are implementing Knowledge distillation, and you have a teacher and a student model. However you simply do not have the necessary GPU resources to fit both the teacher and the student into the GPU at the same time. What is your solution?

@baldwin_IVth Bookmarked it for Sunday reading . Thanks for sharing

pratyakshpatel.github.io/softmax_attent… Back in 7th sem at IISERB, I took a course called Adv Deep Learning, heavy course; in the midsem, there was this one easy attention numerical, adrenaline rushed and I made a stupid mistake on the way, used sigmoid instead of softmax. YES, you read it right. Yeah adrenaline can be harmful at times. since it was a holiday at work, spent some time reading a paper on attention with vs without softmax, wrote this piece as well. Yall would love it.

@Gestalt24 I’m building a tool to automate data flow diagrams. As a data engineer, I spent a lot of time using Figma to manually create animated flows for teams and explanations. It worked, but it was slow and painful. Still a work in progress: dataflow-tau.vercel.app

I'm looking to #connect with people who are interested in: - Build in Public - Full Stack - Startup - Tech - AI - Web3 Let's 🤝 and grow with valuable engagements.

@RealProductGirl I’m building a tool to automate animated data flow diagrams. As a data engineer, I used to rely on #Figma to manually create flow visuals for blogs. It was useful, but incredibly time-consuming. So I decided to build a better way. WIP, but live here: dataflow-tau.vercel.app

I NEED my feed full of builders. What are you working on right now? I don't care if it's a startup or a weekend side project. If you're building something, I want you on my timeline. Reply and let's connect. 👇

@xoaanya Always Claude code : It gets things right faster, needs less back and forth, does better research, writes cleaner code, and stays more consistent when the task gets large or complex.

If you had $20 to invest, which would you choose? – Codex – Claude Code Heard Claude hits limits fast… what would you pick?

This is for #Figma users , check this out medium.com/design-bootcam…

As a DE, I was always drawn to data flow diagrams.I used to create them manually in #Figma & turn them into animated flow but it took way too much time. So I built an app to automate the process.Still a work in progress, but it’s live now: dataflow-tau.vercel.app #buildinpublic

@viberankdev Wow, thank you for sharing . I didn’t know this . Let me add this

@rajvivan Cleaning up AI exports manually is a pain so this is a handy utility. It would be cool if it could handle bulk exports too. Since you are building in public viberankdev could be a great place to list this tool.

A simple tool to clean up and refine your NotebookLM exports for content you own. Try it here: notebook-lm-scrub.vercel.app Would love your feedback 🙌 #buildinpublic #AI #NotebookLM #webapp

@vibeorigin_dev Noted with thanks . Thats a great idea . I am still optimising the performance and will release the next version soon .

@rajvivan This is super useful! One suggestion - you could also add an option to export in different formats (markdown, JSON, etc.) for easier integration with other tools. Great idea building this 🙌

@somi_ai I coded something similar and published it vercel. My app is taking care the PDF and powerpoint watermark generated by NotebookLLM. Check this out notebook-lm-scrub.vercel.app

So my friend made a thing 🍌 Got tired of that Gemini watermark in the corner of Nano Banana images, so he built a little Chrome extension to remove it. It's called "Peel Banana" 🍌 and he just put it up on the Chrome Web Store. Completely free. Chrome Web Store: chromewebstore.google.com/detail/peel-ba…

Time to start something new #newchallenge

My PhD thesis is online! I explore how a classical framework—Bayesian filtering—can tackle modern challenges in online continual learning, bandits, and one-step-ahead forecasting. Enables adaptivity, robustness, and scalability. arxiv.org/abs/2505.07267 grdm.io/posts/bayesian…

@Rushu_Tushu Yes , this book is amazing

Is this book good for beginners who want to learn Artificial Intelligence and Machine Learning?

@Vanarchain @minchoi Beautifully portrayed the future of AI

@minchoi The battle isn’t just about models anymore. It’s about the OS layer, memory, agents, and infrastructure. Whoever controls that backbone controls the future of AI.

Google's AI ecosystem is scary... The next AI war won't be Claude vs ChatGPT. It's who will own the entire AI operating system...

@adaptiveai How to get this … looking forward to test this for my thesis project

Introducing Adaptive Computer. We put AI inside of an always-on personal computer that it uses to get work done. Schedule agents. Create software. Automate anything. As part of the launch, we’re giving one free month of Adaptive to users. Retweet, like, and comment ‘Adaptive’ to get it.

I (finally) put together a new LLM Architecture Gallery that collects the architecture figures all in one place! sebastianraschka.com/llm-architectu…

It's insane to me that despite all the tech progress, you can't find a single half-decent place on the internet to read epics of history like the Mahabharata and Ramayana. Until now. All 10,000 pages. In readable modern English. Free. Read it on grand old books.: grandoldbooks.com/books/mahabhar…