Stock Market Analyzer
End-to-end AI-powered stock analysis platform. Modular ML pipelines with data ingestion, feature engineering, per-ticker model training, and interactive React + TypeScript frontend dashboards.
// Hello, World
Building intelligent systems at the intersection of
software engineering & machine learning
I'm a Computer Science undergraduate and AI/ML Enthusiast passionate about building real systems that ship.
Currently pursuing B.Tech in CSE at IIIT Vadodara – ICD, I thrive at the intersection of algorithms, system design, and machine learning. From building neural networks from scratch to designing full-stack AI pipelines, I love turning complex problems into elegant solutions.
When I'm not coding, you'll find me grinding competitive programming problems, experimenting with LLMs and automation workflows, or exploring cloud platforms.
{
"name": "Shreyash Chaurasia",
"role": "CS Undergrad & ML Enthusiast",
"university": "IIIT Vadodara - ICD",
"interests": [
"Machine Learning",
"Generative AI & LLMs",
"Competitive Programming",
"Cloud & Automation",
"Cybersecurity"
],
"cp_profiles": {
"leetcode": "1519 · 420+ solved",
"codeforces": "1338 · Pupil",
"codechef": "1466",
"atcoder": "1008 · 5 kyu"
},
"coffee": true
}
End-to-end AI-powered stock analysis platform. Modular ML pipelines with data ingestion, feature engineering, per-ticker model training, and interactive React + TypeScript frontend dashboards.
Implemented a neural network framework in pure Python & NumPy — forward/backward propagation, gradient-based optimization, and training loops to deeply understand core ML concepts.
Dockerized workflow automation pipelines using n8n for API orchestration and data processing. Continuously experimenting with AI tools to improve developer productivity.
Database-driven attendance management system with normalized schema design, CRUD operations, and reporting. Features relational data modeling with proper constraints.
A highly aesthetic, premium dark/light themed portfolio featuring physics-based code syntax particles, magnetic hover interactions, dynamic theme ripple transitions, and hacker-glitch typography.
A feature-rich movie tracking and discovery application inspired by LetterBoxd. Currently an ongoing project, actively building out the backend DBMS architecture and integrating cinematic data via REST APIs into a premium UI.
I'm always open to discussing new projects, creative ideas, or opportunities to be part of something amazing. Let's build something great together.
Say Hello