AI & Data Science Engineer | Python | LLM Inference | Fullstack
Second-year B.Tech student in Artificial Intelligence & Data Science at Zeal College of Engineering, Pune. Dedicated to creating high-performance ML pipelines, conversational AI assistants, and modern web applications.
Get to know my academic background, technical focus, and interests.
I specialize in applying Machine Learning, LLM Inference (vLLM), and RAG architectures to practical systems. Passionate about end-to-end development, from data preprocessing and modeling to cloud-ready web frontends.
Technologies, frameworks, and developer tools in my daily workflow.
Real-world AI systems, web applications, and machine learning pipelines.
A comprehensive food delivery platform featuring a real-time business analytics dashboard and an AI mood-based chatbot that offers personalized healthy meal recommendations based on customer emotion.
Hands-on exploration of high-throughput vLLM-based LLM inference. Benchmarks concurrency patterns, PagedAttention KV cache behavior, continuous batching, and GPU memory constraints.
Retrieval-Augmented Generation (RAG) assistant that indexes unstructured documents and facilitates accurate semantic question answering with grounded context and zero hallucination.
Modern Next.js application integrated with Cloudflare Pages for configuring and automating high-deliverability email outreach pipelines with custom dispatch logic.
Sleek, media-rich portfolio website designed specifically for a professional visual effects (VFX) artist, highlighting motion reels, 3D renders, and dynamic client work galleries.
End-to-end machine learning regression pipeline covering exploratory data analysis, feature engineering, model tuning, and an interactive prediction user interface.
Directly loaded in real time via GitHub API from @maddrghule22
Verified credentials and hackathon participation.
HackerRank Certified
HackerRank Certified
Coursera / Stanford Online
Active Participant & Innovator
Have a question, collaboration idea, or an opportunity? Reach out directly.