karananchan,Karan Anchan,github

Hi there! 👋🏽

I'm a computer science grad 🎓, an AI and robotics enthusiast 🤖, and a reader 📚. Constantly diving into the latest and greatest in tech, I also love to stay active and get my game on, be it on the cricket pitch or the football field ⚽🏏.

Here’s a snapshot of what I do, love, and live by:

🛠️ Technical Skills:

Languages: Python, C/C++, Next.js, SQL
Machine Learning Frameworks: TensorFlow, PyTorch
Specializations: Generative AI, NLP, Computer Vision, Reinforcement Learning
Tools: Git/GitHub, MLflow, Power BI

💼 Professional Experience:

AI-driven Education: Designed an ML system at Wizdom Ed that uses knowledge embeddings, NLP, and large language models to personalize learning. We saw a 📈 30% increase in student test scores and a 55% boost in engagement!
Research Innovator: Authored research on text-to-3D generation at Artificer, merging diffusion models with linguistic planning. Our work enhances 3D content creation across VR, visual effects, and product design.

📖 Hobbies & Interests:

Reading: Whether it's the latest in fiction, non-fiction, or science, you’ll find me with a book!
Fitness Buff: Keeping fit with regular workouts 🏋️‍♂️—it’s all about that mind and body connection!
Sports Enthusiast: Cricket and football. There’s nothing like the thrill of the game!

🌟 What Sets Me Apart:

With a unique blend of solid technical skills and real-world AI applications, I love pushing boundaries and solving complex, high-impact challenges 🌍. I bring creativity, energy, and joy to everything I do!

Let's Connect!

I’m always up for a collaboration! Got a question or want to discuss the next big thing in AI? Drop me a message! 📩 Here’s to creating, exploring, and enjoying every moment! 🚀

Karan Anchan's Projects

10_arm_testbed

Explore the 10-Arm Testbed Simulation! 🎲 Utilize Python to test various ε-greedy strategies in a reinforcement learning environment. Visualize and compare agents' performance as they balance exploration and exploitation. Perfect for learners and enthusiasts! 🚀📊

artificer

Artificer generates high-fidelity 3D models from natural language descriptions. It improves on prior methods by using Variational Score Distillation (VSD) for diverse and photorealistic models, and the Recaption, Plan and Generate (RPG) framework to handle complex scenes with multiple objects.

designarium

Designarium empowers designers to navigate a seamlessly interconnected ecosystem. Leveraging the technical prowess of Next.js 13 and TypeScript, alongside the capabilities of Grafbase, Designarium amplifies collaboration, exploration, and the realization of infinite creative possibilities.

detectionalgorithm

Consists of the detection and counting algorithm for yolov7 model

mdp_optimal_grid_sim

Explore the Gridworld Simulation 🌍🚀! An agent navigates a 5x5 grid to maximize rewards, using the Value Iteration algorithm 🔄. Visualizations 📊 show optimal paths and value convergence. Dive into dynamic programming and decision-making! 🤖🧠

transformers_pytorch

Welcome to the Transformer from Scratch repo! 🚀 Explore sequence-to-sequence learning, from input embeddings to attention mechanisms. 🤖 Train, test, and visualize with ease. 📚💡

windy_gridworld_sim

🌍💨 Explore the Windy Gridworld! An agent navigates a 10x7 grid facing wind challenges using Q-learning, SARSA, and Expected SARSA to find optimal paths. Visualize learning stability and compare algorithm performance with running averages and zoomed insights. 🚀📊

Recommend Projects