constcoder={name:'Shreyansh Yadav',skills:['React', 'Node', 'Javascript', 'TypeScript', 'Next.js', 'Express', 'Python', 'Java', 'Spring Boot', 'MySql', 'MongoDB', 'PostgreSQL', 'LLM APIs'],hardWorker:true,quickLearner:true,problemSolver:true,hireable:function() {return(this.hardWorker&&this.problemSolver&&this.skills.length>=5);};};ABOUT ME
Who I am?
My name is Shreyansh. I'm a Full Stack Developer (2026 Batch) specialising in building and shipping production-grade, AI-integrated web applications. I work across the full stack — React.js, Next.js 14, and TypeScript on the frontend; Node.js, Express, and Spring Boot on the backend; MongoDB, MySQL, and PostgreSQL for data. I've independently designed, built, and deployed live AI products integrating LLM APIs (OpenRouter), NLP models (Hugging Face), and voice AI agents (Vapi). GATE 2026 qualified (CS & IT). I'm actively seeking AI-associated or Full Stack SDE roles where intelligent product engineering and shipping real things matter.
Experiences
Skills
PROJECTS
SmartAI Interviewer
constproject={name:'SmartAI Interviewer',tools: ['Next.js', 'TypeScript', 'Node', 'Express', 'MongoDB', 'Firebase', 'JWT', 'OpenRouter LLM', 'Hugging Face', 'Vapi],myRole:Full Stack Developer,Description: Architected and deployed a full-stack AI SaaS platform that conducts real-time, voice-driven mock interviews using a Vapi voice AI agent. Integrated OpenRouter LLM API for dynamic question generation and Hugging Face NLP models for automated answer evaluation and structured feedback scoring. Engineered 10+ RESTful APIs covering interview session management, resume parsing, domain classification, and multi-round evaluation pipeline. Migrated from MERN to Next.js 14 (App Router) + TypeScript, implementing SSR for improved performance and SEO. Secured all routes with JWT middleware and Firebase OAuth.,};PixelMind
constproject={name:'PixelMind',tools: ['React', 'Node', 'Express', 'MongoDB', 'Tailwind', 'Hugging Face', 'Cloudinary],myRole:Full Stack Developer,Description: Built a full-stack AI platform that translates natural language prompts into images using Hugging Face diffusion models, with async request handling for low-latency generation. Designed the AI pipeline layer — prompt sanitisation, API request batching, error/retry handling, and Cloudinary CDN delivery — as a clean, reusable backend service. Implemented a community gallery with pagination, search, and download features.,};VideoTube
constproject={name:'VideoTube',tools: ['Node', 'Express', 'MongoDB', 'JWT', 'Cloudinary', 'Bcrypt', 'Multer],myRole:Backend Developer,Description: Architected a YouTube + Twitter-style production backend with Multer + Cloudinary media pipeline, JWT/Bcrypt auth, and aggregation-based APIs for likes, comments, and subscriptions. Designed optimised Mongoose schemas with indexing supporting full user lifecycle — registration, login, profile, video CRUD, and social interactions. Applied industry-standard patterns: versioned routes, centralised async error handling middleware, and environment-based configuration management.,};AI Chef
constproject={name:'AI Chef',tools: ['React', 'CSS', 'Mistral AI', 'Hugging Face],myRole:Frontend Developer,Description: Built an AI-powered recipe generation web app using React and Mistral AI from Hugging Face. Integrated Mistral's model to generate recipes based on user-input ingredients with proper prompt engineering. Added features to display detailed cooking steps and ingredient breakdown. Optimized for responsiveness and fast loading across all devices.,};PinShare
constproject={name:'PinShare',tools: ['React', 'React-icon', 'React-leaflet', 'Express', 'MongoDB', 'JWT],myRole:Full Stack Developer,Description: Built a full-stack geolocation-based web app using React, Express, and MongoDB. Integrated React Leaflet to enable users to create, view, and manage interactive map markers. Designed a RESTful API for marker CRUD operations with GeoJSON for efficient spatial data handling. Optimized database queries and API performance for a smooth user experience.,};Educations