Narra Suryakoushik Reddy 

AI-powered REST API for real-time dance pose detection and analysis using computer vision.

Key Features:

• Processes videos up to 100MB across MP4, AVI, MOV formats
• Detects 7 distinct dance moves through landmark coordinate analysis
• Extracts 33 body landmarks with timestamped pose sequences
• Deployed on AWS ECS with Docker containerization
• Achieved 99%+ pose detection accuracy with MediaPipe

Technical Highlights:

• Frame-by-frame video analysis pipeline for choreography insights
• Geometric angle calculations for move classification
• Health monitoring and JSON response formatting
• Production-grade deployment on cloud infrastructure

Intelligent resume evaluation system using NLP and ML to match candidates with job descriptions.

Key Features:

• 96% accuracy in resume-job fit prediction
• Semantic similarity analysis using TF-IDF and spaCy
• LanguageTool API integration for grammar validation
• AI-driven suggestions for ATS optimization
• Actionable feedback generation for resume improvement

Technical Highlights:

• Trained classification model with 96% validation accuracy
• NLP-powered semantic matching algorithms
• Grammar checking and readability scoring
• Flask-based REST API for easy integration

High-accuracy facial verification system using Siamese Neural Networks for identity authentication.

Key Achievements:

• 99.6% verification accuracy on test data
• ROC AUC score of 0.998 demonstrating excellent discrimination
• Trained for 1050 epochs using contrastive loss function
• CUDA-accelerated training for faster convergence

Technical Implementation:

• Siamese architecture for one-shot learning
• Contrastive loss optimization for embedding spaces
• Euclidean distance-based similarity computation
• Precision-recall metrics for robust evaluation