I'm a passionate Data Engineer and AI/ML Engineer with expertise in building scalable data pipelines and implementing cutting-edge machine learning solutions. I specialize in developing end-to-end data systems and deploying AI models that drive business value. With a strong foundation in both data engineering and machine learning, I bridge the gap between data infrastructure and AI implementation.

• Data Engineering: ETL pipelines, Data Warehousing, Big Data Processing
• Machine Learning: Deep Learning, Computer Vision, NLP, Model Deployment
• Cloud & DevOps: AWS, Docker, CI/CD, Infrastructure as Code
• Data Analysis: Statistical Analysis, Data Visualization, Business Intelligence

• Technical Achievements:
  • Developed and deployed 5+ production-grade ML models
  • Built scalable data pipelines processing 10GB+ daily
  • Reduced model inference time by 60% through optimization
  • Implemented automated CI/CD pipelines for ML model deployment

A sophisticated deep learning crop recommendation system that provides personalized crop recommendations based on soil conditions and environmental factors.

• Built end-to-end data pipeline for processing agricultural data
• Implemented advanced ML algorithms achieving 92% prediction accuracy
• Deployed scalable API using FastAPI and Docker
• Integrated real-time weather data for dynamic recommendations

Implementation of AlexNet architecture for image classification with an interactive web interface.

• Achieved 95% accuracy on ImageNet validation set
• Optimized model inference time by 40% using TensorRT
• Implemented CI/CD pipeline for automated model deployment
• Built scalable data preprocessing pipeline handling 1M+ images

Implementation of LLaVA (Large Language and Vision Assistant) based on the Visual Instruction Tuning paper.

• Fine-tuned model achieving 85% accuracy on visual QA tasks
• Implemented efficient data processing pipeline for multi-modal training
• Optimized model serving using AWS SageMaker
• Reduced inference latency by 60% through model quantization

Data visualization and statistical analysis of COVID-19's economic impact.

• Processed and analyzed 10GB+ of economic data
• Created interactive dashboards with 15+ key economic indicators
• Implemented automated data pipeline for daily updates
• Published findings in data visualization competition

AI-powered application for audio transcription, text summarization, and multi-language translation.

• Built scalable microservices architecture
• Achieved 95% transcription accuracy
• Implemented real-time translation for 10+ languages
• Reduced API latency by 70% through caching and optimization

NeuroPulse GitHub Repository

A fully on-device, AI-powered focus and energy tracker for iOS 18+ (iOS 26 guidelines). NeuroPulse leverages the latest Apple technologies:

• AppIntents, Widgets, Live Activities, and on-device CoreML
• Privacy-first: all data is local, no network/cloud
• Adaptive SwiftUI 6.0 design, accessibility, and HealthKit integration
• Modern architecture and code organization

Explore the code and architecture in the NeuroPulse repository.