Final-year CSE student building full-stack apps and applied-ML systems — from the API contract to the model weights, with a LeetCode tab open in the background.
I'm a CSE student building end-to-end — type-safe APIs, retrieval pipelines, and the cursor pixels in between. My current obsession is the messy seam between deterministic systems and probabilistic models: agents, RAG, and ML experiments that actually generalize past the toy dataset.
Three years of self-directed building alongside coursework. LeetCode rating north of 1900, 20+ ML/DL models trained on real data, and a stack of full-stack side projects I keep iterating on. Aiming at SDE and ML roles for 2026.
The boring answer to 'what do you use?' — grouped by where they live in the stack.
A few systems I've designed and built end-to-end. Hover for the back-of-the-napkin story.
A self-building knowledge base and full-stack RAG pipeline that ingests URLs and documents to provide LLM-generated answers grounded in citations. Features a custom data pipeline with Playwright scraping, ChromaDB vector search, and an interactive 2-D PCA retrieval map to visualize the embedding space.
A Python-based CLI tool and library that automatically converts Swagger and OpenAPI specifications into visual Draw.io diagrams. Features a custom graph engine and layout strategies to instantly map complex API endpoints, data models, and architectural relationships.
A full-stack asynchronous task scheduling and execution system featuring custom-engineered worker pools. It enables defining, scheduling, and monitoring concurrent processes with real-time state updates via WebSockets.
A deep learning web application that classifies brain MRI scans into four distinct categories (Pituitary, Meningioma, Glioma, or No Tumor). Leverages a fine-tuned VGG16 Convolutional Neural Network deployed via a Flask and Bootstrap interface.
Formal schooling plus the self-directed track I'm running now.
Coursework spans DSA, OS, DBMS, computer networks, and machine learning. Final-year focus on applied ML and full-stack systems. CGPA: 8.5/10 (up to 6th semester).
Things I'm proud of from the last few years — from competitive programming milestones to multiple projects.
Pushed past the 1900 contest rating mark on LeetCode after 700+ problems solved across DP, graphs, and segment trees.
Secured a top 10 finish in Inter-College Hackathon, building a AI-powered Resume Evaluator. Evaluated on innovation, technical complexity, and impact.
Trained and shipped 20+ ML/DL models across vision, NLP, and tabular data — from CNN classifiers to small fine-tuned transformers.
Built and deployed multiple end-to-end applications using React, Django/FastAPI, authentication systems, databases, and cloud deployment workflows.
Vendor certs I've earned to back the system architectures I sign off on.
Have a system in mind, a half-formed idea, or just a hard problem? I read every message.
Open to SDE / ML roles for 2026, internships, and interesting side collabs.