Available for opportunities

Abdullah
Khan

AI / ML Engineer

Final-year CS student building production-grade machine learning systems. From data pipelines to deployed APIs - I build AI that actually ships.

View Projects → Download Resume Get In Touch
3+
Live Projects
5+
Frameworks
31
Certifications
Scroll
Python ·
Machine Learning ·
Deep Learning ·
Generative AI ·
LangGraph ·
CrewAI ·
FastAPI ·
Streamlit ·
MLflow ·
HuggingFace ·
Scikit-learn ·
NLP ·
RAG Applications ·
Vector Databases ·
Python ·
Machine Learning ·
Deep Learning ·
Generative AI ·
LangGraph ·
CrewAI ·
FastAPI ·
Streamlit ·
MLflow ·
HuggingFace ·
Scikit-learn ·
NLP ·
RAG Applications ·
Vector Databases ·
Who I Am

Building intelligent
systems that ship.

Abdullah Khan

Abdullah Khan

AI/ML Engineer

I'm Abdullah Khan, a final-year Computer Science student at The University of Faisalabad with a CGPA of 3.57/4.0. I specialize in the full AI/ML lifecycle — from raw data to production-deployed models.

My work spans supervised & unsupervised learning, NLP, Generative AI, and MLOps. I've built and deployed multiple production-grade ML applications on Hugging Face Spaces, integrating REST APIs, model serialization, and clean frontend interfaces.

I hold certifications from DeepLearning.AI and IBM covering deep neural networks, agentic AI frameworks (LangGraph, CrewAI, AutoGen), and vector databases for RAG applications. I'm passionate about writing clean, production-ready code that solves real-world problems.

GitHub Activity & Current Focus

GitHub Graph
🎓
Education
BSc Computer Science · The University of Faisalabad · Expected May 2026
📍
Location
Faisalabad, Punjab, Pakistan · Open to remote opportunities worldwide
🎯
Focus Areas
MLOps · Generative AI · Agentic AI · NLP · Production ML Deployment
⚡
Active Project
Building an MLOps pipeline for Fraud Detection System with FastAPI, MLflow & Gradio
Top Languages
Jupyter 52%
TypeScript 34%
HTML 7%
Python 4%
CSS 1%
JS 0%
Currently Exploring
  • → Agentic Workflows
  • → Advanced RAG & Vector DBs
  • → MLOps & Local LLMs
  • → Fine-tuning Large Language Models
  • → Cloud-Native AI Infrastructure
What I Know

Technical Skills

🐍 Languages & DBs
Python SQL MySQL JavaScript C++ Java PHP
🤖 Machine Learning
Scikit-Learn XGBoost Random Forest SVM Supervised Learning Unsupervised Learning Feature Engineering
🧠 Deep Learning & Vision
PyTorch TensorFlow Keras Hugging Face CNN & RNN Deep Neural Networks NLP
✨ Generative AI & LLMs
LangChain LangGraph CrewAI AutoGen BeeAI RAG Applications Prompt Engineering Vector Databases
📊 Data Science
Pandas NumPy Matplotlib Data Cleaning Exploratory Data Analysis Data Visualization
🚀 MLOps & Web Apps
MLflow FastAPI Flask Streamlit Gradio REST APIs Git
What I've Built

Projects

01

🎓 UniMind — RAG University Study Assistant

⭐ Featured
Live

Ask questions about your university notes using AI. Features Additive Vector Indexing, Smart Multi-Format OCR, and Resilient LLM Routing.

  • Additive PDF Ingestion with Smart Multi-Format OCR (PyMuPDF & Tesseract)
  • Resilient LLM Routing with fallback mechanisms (gpt-oss-120b & Qwen2.5-72B)
  • Stateless Session History & Premium Claude-Style UI with LaTeX Support
Python Streamlit LangChain ChromaDB HuggingFace
🚀 Live Demo { } GitHub
02

🛡️ Fraud Detection System
Live

A real-time, event-driven fraud prevention MLOps system processing streaming transactions with Apache Kafka, FastAPI, and Gradio.

  • End-to-End MLOps pipeline from data ingestion to model training
  • Event-driven architecture streaming transactions via Apache Kafka
  • Dual API interfaces: FastAPI for developers and Gradio for UI testing
  • Experiment tracking via MLflow for model metrics (ROC-AUC)
Python Kafka FastAPI Gradio Docker MLflow Scikit-Learn
🚀 Live Demo { } GitHub
03

💎 Diamond Price Prediction
Live

A production-grade MLOps system for predicting diamond prices with a modular pipeline tracked via MLflow.

  • End-to-end MLOps pipeline: ingestion → transformation → training → deployment
  • MLflow experiment tracking comparing multiple regression models
  • FastAPI REST endpoint with Pydantic v2 input validation
  • Gradio frontend with structured 4C diamond grading reference guide
Python FastAPI MLflow Gradio XGBoost
🚀 Live Demo { } GitHub
04

🎬 CineMatch
Live

Content-based recommendation engine using cosine similarity across 5,000+ movies with real-time posters.

  • CountVectorizer feature extraction on genres, cast, keywords & overviews
  • TMDB REST API integration for live posters & release years using movie IDs
  • Cached model inference with st.cache_resource for fast cold starts
Python Scikit-learn Streamlit TMDB API
🚀 Live Demo { } GitHub
05

🏥 HealthGuard AI
Live

Three independent ML models predicting Diabetes, Heart Disease, and Parkinson's Disease.

  • 3 separate SVM/Logistic Regression models with dedicated StandardScalers
  • 1,266 patient records spanning 8–22 clinical features per disease
  • Color-coded results with responsible AI disclaimers throughout
Python Scikit-learn Streamlit NumPy
🚀 Live Demo { } GitHub
Credentials

Certifications

🤖
IBM
IBM RAG and Agentic AI Specialization
✨
IBM
Generative AI for Data Scientists Specialization
🧠
DeepLearning.AI
Deep Learning Specialization
📊
DeepLearning.AI
Machine Learning Specialization
🐍
University of Michigan
Python for Everybody Specialization
Education
🎓
Bachelor of Science in Computer Science
The University of Faisalabad · Punjab, Pakistan
Expected Graduation: May 2026
⭐ CGPA: 3.57 / 4.0
Let's Work Together

Get In Touch

I'm actively seeking internships and entry-level opportunities in Data Science, Machine Learning, and AI Engineering. If you're building something intelligent and need someone who can ship — let's talk.

✉️
Email
abdullahkhansherwani09@gmail.com
 💼
LinkedIn
linkedin.com/in/abdullahks-devhub
 💻
GitHub
github.com/abdullahks-devhub
 🤗
HuggingFace
huggingface.co/spaces/AbdullahKS-Devhub

Open to Opportunities

Final-year CS student graduating May 2026. Available for internships, part-time roles, and freelance ML projects. Based in Pakistan, open to remote work globally.

Send Me a Message →
Email Copied! 🚀