Hi, I'm Amit Dhakad
Full-Stack & GenAI Engineer
B.Tech Graduate in Electronics & Communication Engineering from NIT Bhopal (MANIT). Skilled in building MERN web applications, Python pipelines, and generative AI systems.
Full-Stack & GenAI Engineer
B.Tech Graduate in Electronics & Communication Engineering from NIT Bhopal (MANIT). Skilled in building MERN web applications, Python pipelines, and generative AI systems.
I am a B.Tech Graduate in Electronics & Communication Engineering from NIT Bhopal (MANIT) with a strong passion for software engineering, web development, and Generative AI. I specialize in building MERN stack applications, Python data pipelines, and intelligent Generative AI systems.
My experience spans full-stack web development, prompt engineering, agentic orchestration, and systems optimization. I enjoy building automated validation pipelines, building recommendation engines, and solving complex problems with over 500 DSA solutions.
Maulana Azad National Institute of Technology (MANIT), Bhopal
Jawahar Navodaya Vidyalaya, Sheopur
Score: 94.8%
Jawahar Navodaya Vidyalaya, Sheopur
Score: 92.0%
DSA Solutions
Projects Completed
Events Coordinated
Built a 15M-parameter transformer model from scratch using PyTorch, handling the tokenization and self-attention layers from the ground up. Used Pandas to structure datasets, and managed weight initializations via NumPy. Cut training times using mixed-precision and CUDA optimization.
Developed a Python & Gemini LLM app to automate real-time medical claim validation and fraud detection. Designed rigorous validation test cases and evaluation criteria to achieve 90%+ verification accuracy, implementing automated compliance flags.
Built a multi-tiered marketplace platform featuring an LLM-driven application for real-time agricultural diagnostics. Experimented with prompt architectures to optimize user interaction and backend decision-making, and simulated load testing for concurrency.
Built a content-based recommendation engine utilizing NLP-driven insights from unstructured audio data. Transformed metadata into actionable user preference signals for real-time recommendations, running on a low-latency decision system.
A feature-rich food delivery platform with real-time order tracking, secure JWT authentication, and role-based dashboards (Admin/User). Optimized REST APIs and MongoDB queries, reducing response time by 40%.
A multi-user whiteboard for live collaboration. Built a custom stack-based Undo/Redo system for state synchronization across concurrent users, enabling seamless real-time collaboration.
A comprehensive video streaming platform mimicking YouTube's core features. Built with React.js, featuring video playback, search functionality, trending videos, and user playlists.
A normalized (3NF) 7-table relational database for hospital records. Enforced strict data integrity using primary/foreign key constraints, and wrote advanced SQL queries for performance reports.
I'm always open to discussing new projects, creative ideas, or opportunities to be part of your vision.