Open to Work
Muhammad Rizwan
AI/ML Engineer — Lahore, Pakistan
I build multi-agent systems, computer vision, and generative AI products — end to end, from research to production.
I build multi-agent systems, computer vision, and generative AI products — end to end, from research to production.
AI/ML Engineer and BS Artificial Intelligence graduate from The Islamia University of Bahawalpur (2022–2026).
I build practical AI systems across machine learning, NLP, computer vision, and generative AI—from model development to API-based applications. My work includes bilingual scam detection, mango-leaf disease classification, pose-based fitness coaching, and the MARS-NLP research project on multilingual scam-message classification.
My engineering toolkit includes Python, scikit-learn, TensorFlow/Keras, PyTorch, OpenCV, FastAPI, Flask, Docker, and Git/GitHub. I’m interested in entry-level AI/ML engineering roles, internships, contract work, and remote opportunities.
Research project studying multilingual scam-message classification with classical ML baselines and transformer-model comparisons. The public repository includes the project code, documented results, release materials, and manuscript files. This work is a research project and is not presented as a published paper.
Bilingual English and Roman Urdu scam/fraud message detection using TF-IDF and Random Forest, with a Flask application and SQLite-backed history. The project focuses on SMS and messaging text.
Final-year project for mango-leaf disease classification using transfer learning with MobileNetV2. The project reports 92.1% accuracy across nine disease/health categories and uses a Flask-based application.
Conversational assistant project with a FastAPI backend, chat history, mood/emotion signals, and support planned for English, Urdu, and Roman Urdu interactions.
AI fitness-coaching application combining pose estimation, repetition counting, and live feedback to support exercise-form analysis.
AI-powered learning project using the Gemini API to support educational content discovery, learning roadmaps, and progress features.
Custom object-detection project for identifying mobile phones in images or video frames using a fine-tuned YOLO model.
Completed internship assignments and capstone work; project deliverables were accepted. Recommendation and evaluation documents were received.
AI/ML learning track covering machine learning, deep learning, mathematics, and applied project work, including the GuardAI bilingual scam-detection capstone.
Graduated in Artificial Intelligence. Final-year project: MangoLeafScan AI.
Only credentials and learning programs currently documented in my records are listed here. Additional certificates can be added with their verified titles and links.
Silver-tier participation in the AI Seekho learning program.
Released project code, experiment documentation, and manuscript materials for a multilingual scam-message classification research project. Journal publication status: not yet published.
Open to full-time, contract, and remote AI/ML roles — Pakistan and internationally.