Sai Bhargav Damarla

About

Computer Science graduate with hands-on experience in Machine Learning, Deep Learning, NLP, and Generative AI. Experienced in building and evaluating Transformer, GAN, and YOLO-based models using Python, TensorFlow, PyTorch, and Hugging Face.

Education

• B.Tech in Computer Science and Engineering
  SRM University, Andhra Pradesh
  Computer Science and Engineering · 2022 – 2026
• Intermediate (MPC)
  FIITJEE Junior College
  MPC · 2020 – 2022
• SSC
  FIITJEE International School
  2020

Skills

• C <3 months
• Computer Networks <3 months
• Deep Learning <3 months
• ChatGPT <3 months
• Python <3 months
• Machine Learning <3 months
• Java <3 months
• JavaScript <3 months
• NumPy <3 months
• Pandas <3 months
• scikit-learn <3 months
• OpenCV <3 months
• PyTorch <3 months
• TensorFlow <3 months
• Operating Systems <3 months
• CNN (Convolutional Neural Network) <3 months
• Yolo <3 months
• Natural Language Processing <3 months
• LLMs <3 months
• Transformers <3 months
• Data Structures and Algorithms <3 months
• Oop <3 months
• Generative Adversarial Networks (GANs) <3 months
• Generative AI (ChatGPT, Copilot, Gemini) <3 months

Tools / apps / platforms

• Git <3 months
• GitHub <3 months
• Hugging Face <3 months

Projects

• Image Generation Using DCGAN
  Python, TensorFlow

  Developed and trained DCGAN models on MNIST, SVHN, and CelebA datasets for synthetic image generation. Designed custom generator and discriminator architectures optimized using Adam optimizer and Binary Cross Entropy loss. Achieved high-quality synthetic image generation after 100+ training epochs, validated through visual evaluation.

• Real Time Object Detection in Aerial Surveillance Using YOLO Models
  Python, PyTorch, YOLO Models

  Implemented and trained multiple YOLO architectures (v8, v10, v11) for real-time object detection on aerial imagery using the VisDrone2019 dataset. Conducted comparative performance analysis using mAP@0.5, mAP@0.5:0.95, precision, and recall to evaluate detection accuracy. Identified YOLOv11 as the most effective model for detecting small and densely packed objects in aerial surveillance scenarios.

• TITAN: Triple-Stream Integrated Attention Network for Title-Based Fake News Detection
  Python, NLP

  Developed a Transformer-based model for title-only fake news detection using semantic, emotional, and syntactic attention streams. Integrated a pre-trained DeBERTa-v3-base encoder with gated multi-view fusion and diversity-aware regularization. Fine-tuned the model using PyTorch and Hugging Face Transformers, with NumPy, Pandas, and Scikit-learn for preprocessing and evaluation. Achieved 85.0% accuracy on GossipCop and 96.41% on WELFake.

Courses & certifications

• MongoDB · SmartBridge
• Google AI Essentials · Google (Coursera)
• The Joy of Computing Python · NPTEL
• Cyber Security · Google (Coursera)