Skip to content
View H0NEYP0T-466's full-sized avatar
:shipit:
:shipit:

Highlights

  • Pro

Block or report H0NEYP0T-466

Block user

Prevent this user from interacting with your repositories and sending you notifications. Learn more about blocking users.

You must be logged in to block users.

Content in all repositories owned by your account will be closed.
Maximum 250 characters. Please don’t include any personal information such as legal names or email addresses. Markdown is supported. This note will only be visible to you.
Report abuse

Contact GitHub support about this user’s behavior. Learn more about reporting abuse.

Report abuse
H0NEYP0T-466/README.md

Typing SVG


h0neyp0t-466


📌 ABOUT ME

Hey, hope you're doing well! I'm Fezan, an AI engineer whose primary hobby is building things and token eating (once burned 312M+ tokens in a single 4-hour sprint). Always open to collaborating on research papers, interesting projects, or wild ideas. Drop me an email or text anytime.

  • 🎓 Education: Bachelor of Computer Science @ Lahore Garrison University
  • 🔬 Core Research: Medical imaging pipelines (classification, segmentation, foundation models) and agentic workflows.
  • 📂 The Repositories: You'll find a wide spectrum of work here, from early practice scripts, simple todo lists, and games, all the way to custom language compilers, OS process managers, from-scratch GPT implementations, and multi-organ medical AI platforms.

⚡ TECH STACK

🤖 Machine Learning & Deep Learning

Python PyTorch ML/DL RAG Pipelines LoRA / QLoRA Local LLMs Multi-Agent Systems

🌐 Full-Stack (MERN + FastAPI)

FastAPI React Node.js Express MongoDB TypeScript C++

☁️ Cloud & Infrastructure

Docker AWS Vercel PostgreSQL


🎯 FEATURED PROJECTS

OpenMed (AI Hospital)

Multi-Organ Disease Classification & Segmentation

Full-stack clinical diagnostic system supporting 10+ organ systems for classification and segmentation.

  • Tech: PyTorch, FastAPI, React, TypeScript, Three.js
  • Assistive: Prompt-based interactive segmentation, automated report generation, AI diagnostic assistant, 3D anatomical organ models

GPT-2

From-Scratch Transformer Architecture & Engine

Lightweight from-scratch reimplementation of OpenAI's GPT-2 (124M) with real-time text streaming.

  • Tech: PyTorch, FastAPI, React, TypeScript, Transformers
  • Features: Causal self-attention, BPE tokenization, FastAPI streaming backend, interactive React inference UI

BTSC-UNet-ViT

Brain Tumor Classification & Segmentation

Hybrid vision transformer and convolutional network for joint multi-class diagnosis and lesion segmentation.

  • Tech: PyTorch, ViT, U-Net, Medical Imaging
  • Benchmark: 99.26% test accuracy across 90,000+ medical images (Shipped)

DINOv2-ISIC

Dermoscopic Skin Lesion Benchmark

Self-supervised vision foundation model fine-tuned on dermatoscopic image benchmarks.

  • Tech: PyTorch, DINOv2, Computer Vision
  • Benchmark: SOTA F1 Macro 0.7139 on ISIC 2019 (Beat official leaderboard)

📊 GitHub Analytics

GitHub Streak

GitHub Stats

Activity Graph

🌐 Connect With Me

LinkedIn Stack Overflow Email


Snake animation

Made with 🤍

Pinned Loading

  1. BTSC-UNet-ViT BTSC-UNet-ViT Public

    🧠 BTSC-UNet-ViT— a🚀brain tumor segmentation & classification system combining 🧩Vision Transformers (ViT) for tumor detection and 🎯 UNet for precise pixel-level segmentation from MRI scans. Features…

    Python 5

  2. Dinov2-ISIC Dinov2-ISIC Public

    Skin lesion classification with a DINOv2-B backbone (86M params) on ISIC 2019 (8 classes). Self-contained Colab training script + FastAPI inference backend + React/TypeScript frontend. End-to-end, …

    Python 3

  3. GPT GPT Public

    🤖 From-scratch GPT in PyTorch — bigram, single-head & multi-head attention • FastAPI /generate endpoint • React + Vite frontend • Character-level tokenizer • CUDA support

    Python 3

  4. GPT-2 GPT-2 Public

    A minimal GPT-2 implementation from scratch — FastAPI backend for text generation/training + React + TypeScript frontend for an interactive experience

    Python 4

  5. finetuneLLM finetuneLLM Public

    💻⚡ Terminal-style chat interface for local LLMs using React + TypeScript & FastAPI. Supports GGUF models, real-time token streaming, MongoDB chat history, and LoRA-based fine-tuning of Microsoft Ph…

    Python 4

  6. Pen2PDF Pen2PDF Public

    ⚡ Pen2PDF Suite – an all-in-one 🚀 productivity platform ✨ with 🤖 AI-powered text extraction (PDF/Images → Markdown 📝), 📅 smart timetable management (CSV/Excel import 📊), ✅ todo lists with subtasks📈…

    JavaScript 10 2