TA

Thushar P A

Computer Science undergraduate · Machine Learning & Software Development

Changaramkulam, Kerala, India

@thushar_p_a

0 followers

About

Fourth-year Computer Science undergraduate with hands-on experience across the full software and ML development lifecycle, including model development, backend integration, and deployment. Built end-to-end projects in medical imaging, real-time computer vision, and SQL-based analytics pipelines.

Experience

  • AI/ML Intern
    ICT Academy of Kerala

Education

  • BTech
    Government Engineering College Kozhikode
    Computer Science and Design · 2023 – 2027

Skills

  • SQL <3 months
  • FastAPI <3 months
  • Python <3 months
  • Model Evaluation <3 months
  • NumPy <3 months
  • Pandas <3 months
  • scikit-learn <3 months
  • OpenCV <3 months
  • Flask <3 months
  • PyTorch <3 months
  • TensorFlow <3 months
  • CNN (Convolutional Neural Network) <3 months
  • Yolo <3 months
  • ETL <3 months
  • Object Detection <3 months
  • SQLAlchemy <3 months
  • Image Classification <3 months

Tools / apps / platforms

  • Docker <3 months
  • Git <3 months
  • GitHub <3 months
  • Linux <3 months
  • ONNX Runtime <3 months
  • PostgreSQL <3 months
  • Render <3 months
  • Vercel <3 months

Projects

  • Python, PostgreSQL, SQL, SQLAlchemy

    Solo project building a Python ETL pipeline to transform ball-by-ball JSON data into analytics-ready PostgreSQL tables and advanced SQL analytics for batting positions, match phases, bowling, and team performance.

  • Unblinder
    Python, YOLOv11, ONNX Runtime, OpenCV, FastAPI, WebSockets, Groq API, OpenStreetMap

    AI navigation assistant for the visually impaired. Built a real-time walking assistant combining object detection, live camera input, routing, spoken navigation briefings, scene summaries, voice assistance, and weather information; deployed live on Render.

  • Python, YOLOv8, OpenCV, Flask, React, Node.js, MongoDB

    AI-based bone fracture detection platform. Collaborated in a 3-member team, owning the ML pipeline and backend integration; achieved 86% precision, 74% recall, and 80% F1-score detecting bone fractures across 4,000+ X-rays using a custom YOLOv8 pipeline.

Courses & certifications

  • Python for Data Science (Elite) · NPTEL, IIT Madras
  • Getting Started with Generative AI · IBM
  • Google Data Analytics Professional Certificate · Google & Coursera

Explore the Broxer community

🤖
Online · instant AI help
Broxer