Just a developer trying to make life a little more “hands-free.” ✋✨ I built a gesture-controlled mouse using Python and MediaPipe. It wasn’t easy—filtering out hand jitters and detecting a reliable “pinch” in real time took a lot of trial and error. The current setup: 🖱️ Pinch = Left Click 📜 Two-finger lift = Scroll 👍 Thumbs up = Volume Up Still experimenting and improving, but turning an idea into something that actually works feels great. You can also download the .exe file from the Releases section and try it directly. It’s simple, it’s experimental, and I’m learning something new with every line of code 📈 Try it here 👇 👉 https://lnkd.in/g9sZUCjM #CodingLife #SoftwareEngineering #AI #HandsOnLearning #PythonProject #Innovation #BuildInPublic

Built a GenAI application that converts YouTube videos into structured articles, downloadable PDFs, and responsive webpages. The system uses a multi-step pipeline: transcript extraction, content cleaning, article generation, and multi-format output creation. Tech stack: Python, Streamlit, LangChain, Groq (LLaMA 3.3), FPDF This project helped me understand how to design end-to-end GenAI workflows beyond simple summarization. 🔗 GitHub: https://lnkd.in/gygZUgGG #Innomaticsresearchlabs #GenerativeAI #Langchain #Python #Datascience

My new course on ActionVFX is coming out soon: Python & AI Integration for Nuke. I really wanted to make this the ultimate and most up to date guide for Nuke artists, so we're diving deep into python and programming concepts and build real-world tools together - zero programming knowledge required. And we will also build real world AI tools without any disturbing AI Hype: An AI Chatbot to use right inside Nuke and an AI Inpaint Edit Gizmo. Time-limited pre-orders are live now: https://bit.ly/3ORaa88

Day 16 of My AI Journey 🚀 Today I focused on using Python to automate simple tasks. Covered: 👉 Writing small automation scripts 👉 Working with files and data together 👉 Improving efficiency by reducing manual steps What I worked on: 👉 Built scripts to process files and organize data automatically 👉 Practiced combining multiple concepts into a single workflow Key takeaway: 👉 Automation is where programming starts delivering real value This step is helping me connect fundamentals with practical, real-world use cases. #Python #AI #LearningInPublic #BuildInPublic

Solved Kth Smallest Element in a BST today Used an iterative inorder traversal (DFS with stack) to efficiently find the answer without extra space for storing nodes. Key idea: 👉 Inorder traversal of a BST gives nodes in sorted order 👉 So the k-th visited node is the answer Instead of recursion, I used a stack to simulate traversal: Go left as much as possible Process node Move right Clean, efficient, and interview-friendly ✅ Time: O(H + k) Space: O(H) (stack) Consistent practice is making these patterns feel natural 🚀 #LeetCode #DataStructures #Algorithms #Python #CodingInterview

🚀 Built my first RAG (Retrieval-Augmented Generation) Chatbot using Python! Instead of guessing, this chatbot reads, understands, and answers directly from custom data 📄➡️🤖 Powered by FAISS, HuggingFace embeddings, and Groq LLM, it delivers fast and context-aware responses. 💡 From static text → to intelligent conversations This is a small step into the world of AI-powered applications, but a big leap in how machines interact with knowledge. #AI #MachineLearning #LangChain #Python #RAG #GenAI #DataScience

Rust-Based Type Checkers Outperform Mypy and Pyright by Order of Magnitude 📌 Rust-powered Python type checkers like Pyrefly and ty now slash analysis time by 14x, outpacing Mypy and Pyright in speed - a game-changer for CI/CD and AI IDEs. Their low-level Rust core delivers massive performance gains while tackling long-standing tooling bottlenecks. Developers can finally enjoy blazing-fast typing without sacrificing flexibility or accuracy. 🔗 Read more: https://lnkd.in/dbaHD8Yp #Pyrefly #Ty #Mypy #Pyright #Rust

I just dropped a new video breaking down my Personal Assistant project. In 6 minutes I walk through the full system — architecture, code, RAG memory layer, and a live demo. If you're into Python, FastAPI, or AI automation, this one's for you. 👉 https://lnkd.in/dtKqdMiM #Python #FastAPI #AI #RAG #BuildInPublic

Day 9 of My AI Journey 🚀 Today I focused on strengthening problem-solving and writing more structured Python code. Covered: 👉 Functions for modular code 👉 Combining loops and conditions 👉 Improving code readability and structure 👉 Basics of thinking in terms of inputs → processing → output Built: 👉 Refined previous programs with better structure and reusability Key takeaway: 👉 Strong fundamentals in programming are essential before building advanced AI systems #Python #AI #LearningInPublic #BuildInPublic

New video out today — this time on the PyCharm channel! We built a TensorFlow model from scratch, step by step, using a Jupyter notebook inside PyCharm. The goal was to make it genuinely beginner-friendly: no assumptions, no hand-waving, just actual code that runs. What we covered: — Loading and visualising a real dataset (fashion images — much more fun than MNIST) — Building and comparing two different model architectures — Evaluating accuracy and actually understanding what the number means — Digging into where the model gets confused and why (spoiler: shirts and pullovers are hard) — Using PyCharm's AI assistant to speed up the parts that don't need to be slow One thing I always try to do in these videos: show the thinking, not just the result. Why do we normalise the pixel values? Why ReLU? Why does the second model not actually justify its extra training time? If you're getting started with TensorFlow or just want to see how a clean ML workflow looks inside a proper IDE — this one's for you. 👉 https://lnkd.in/eAXj8K-F #TensorFlow #MachineLearning #Python #PyCharm #JetBrains JetBrains