Aman Agrawal

About

Data science and computer vision enthusiast with hands-on experience building end-to-end AI projects using Python, SQL, Scikit-learn, PyTorch, YOLOv8, and OpenCV. Experienced in data preprocessing, model training and evaluation, object detection, OCR workflows, and real-time computer vision applications.

Education

• B.Sc. (Hons.) Physics
  Magadh University
  Physics · 2022 – 2025

Skills

• SQL
• Data Preprocessing
• Python
• Classification
• Data Cleaning
• Model Evaluation
• NumPy
• Pandas
• scikit-learn
• OpenCV
• PyTorch
• Supervised Learning
• Random Forest
• Exploratory Data Analysis
• Object Detection
• Data Labeling
• Regression
• CVAT
• Train test split

Projects

• Delhi Air Quality Prediction
  Python, Pandas, NumPy, Scikit-learn, Random Forest

  Built a machine-learning pipeline using Delhi air-quality data from multiple monitoring-station files. Performed data loading, preprocessing, feature preparation, training, and evaluation using Random Forest.

• SmartTrafficAI — Emergency Vehicle Detection & Traffic Signal Automation
  Python, YOLOv8, OpenCV, EasyOCR

  Developed a real-time emergency-vehicle detection pipeline for ambulance, police vehicle, firetruck, and car classes. Prepared and merged custom datasets, performed image labeling and quality cleanup, and trained YOLO models for traffic-scene detection.

• RoadGuard AI — Real-Time Traffic Safety Monitoring
  Python, YOLOv8, OpenCV, EasyOCR

  Built a real-time computer vision system for helmet detection, vehicle/number-plate detection, and traffic-violation monitoring. Integrated license-plate cropping and OCR workflow with real-time detection display and evidence screenshot saving.