Tekraj Awasthi🧑💻🇳🇵
@trawasthi_ai
MS AI @UniFAU• R&D Insights on AI (RL & LLM)🚀• Simplifying ML and Maths • MLOps • Ex- Senior DE @IncedoInc
Mathematics for Machine Learning and Deep Learning 🚀🤖 : A Complete Roadmap ⤵️ #machinelearning #deeplearning #mathematics #artificialintelligence #GenAI #OpenAI #LlamaRace #100daysofcode
Always saying this! Build End-to-End ML systems.🚀 For that, upskill in - MLOps(MLFlow, DAGHubs, DVC, etc.), - Flask, FastAPI, - Graph DB, Vector DB(Pinecone) in high demand - Docker, - Streamlit/Gradio, etc.
Unless you're interested in ML research, you don't need to spend too much time exploring the math and architectures of every ML and LLM algorithm. Focus on learning how to build real-world ML systems, the trade-offs of different algorithms, and their engineering side.
It's sad, but this is the current state of AI (Cursor): Sometimes it works. And when it doesn't work, no amount of prompting will get you out of the rabbit hole. You either know what you are doing, or you are stuck forever. #Copied from @svpino.
This is next-level surrender by AI😱🤯 Hard to believe, but yes, Cursor Pro(not even a free tier) with Gemini2.5 Pro(considered one of the best) gave up and how. Let's all laugh again at those AI influencers who say "AI will completely replace developers."😂
Returning to X after a week-long break. Did I miss anything crazy here?
Even Dinosaurs can return, but not the IT job market of 2021/2022. That was the prime job market.
Wasn't expecting this many people will ignore that '!' in my last post. Even the CS/IT guys also...😭 BTW, Sorry to all the folks in my inbox asking for Meta Interview roadmap. It was just trend experiment. Hopefully we'll manifest it together some day.🙂
Not able to understand Machine Learning and Deep Neural Networks well?🤔 Here are 6 brilliant visualizations of the ML & Neural Network models, including 3D viz. of LLM 🚀🧵
Mamba is a new neural network architecture for sequence modeling, using selective state space models (SSMs) to handle long sequences efficiently, in which Transformer architecture struggles. It offers linear time complexity, whereas transformer-based models have quadratic…
An African Proverb that not many know but is life-changing: As you enter your 20s, if you continue to believe in yourself through ups and downs, you'll surely be 25 after five years. #WednesdayMotivation
🚨RIP Cursor/Windsurf. You no longer need to leave VS Code to take help from AI copilots. In case you're still switching b/w VS Code and ChatGPT/Claude/Gemini while coding, you're a retarded dev. Anthropic published the Claude Code extension 3 weeks ago. Just install it!
AI scolding itself after being miserably poor for hours on finding the bug created by itself.😂
Evolution of Attention Mechanisms and Transformer Architectures🤖🚀 Here's everything you need to know about the 🧵 - Pre-Attention Era, - How Attention Mechanism came, - Strengths & weaknesses of different variants, - Transformer Architectures and their variants 1/6
That's just a diagram. How can he get an orgasm looking at it? Meanwhile, the diagram:
Another brilliant Transformer Architecture Explainer with Visualization📇🚀 - by Polo Club of Data Science @ Georgia Tech
PyTorch in One Hour🤖 A gem of a blog by @rasbt for you to learn all the essential concept of tensors and their usage in PyTorch🚀 Link: sebastianraschka.com/teaching/pytor…
5 levels of Machine Learning projects - Covers the project level and skills needed for entry-level, Mid, and senior-level ML expertise. 🚀 If you're a beginner or an intermediate-level ML practitioner, you simply can't miss watching this video by Marina Wyss. Thank me later!…
