Supradeep Danturti Email & Phone Number

Toronto, On, Ca

Toronto, On, Ca

Toronto, Ontario, Canada

In my role as a member of the Release Management Team:• Engineered a complex data visualization dashboard for cybersecurity and fraud prevention projects using Grafana, Kibana, and Python to monitor millions of daily transactions, resulting in a 20% decrease in deployment errors.• Developed and implemented a forecasting algorithm with 92% accuracy using deep learning techniques such as RNNs and LSTM networks to predict future trends for strategic decision-making purposes, and used Apache Kafka + ELK Stack to handle real-time data streams• Automated document creation processes with SQL and Python, streamlining daily monitoring procedures and improving productivity by 15% while reducing manual labor by 30%• Utilized Python scripts to automate ETL processes, resulting in a 50% reduction in manual data processing time and increased daily data monitoring efficiency by 60%.

Noida, Uttar Pradesh, India

As a key member of the Product R&D Team, I actively contributed to the development of two cutting-edge projects:• Developed a machine learning algorithm using Python and Node.js to analyze employee attendance data in real-time, resulting in a 20% increase in accuracy compared to manual tracking methods.• Proposed data-efficient techniques: data augmentation, transfer learning, and knowledge distillation to design and implement a face recognition model using deep learning algorithms (CNNs) with TensorFlow, Keras and other Deep Learning techniques. Achieved a 20% increase in model efficiency compared to previous iterations.• Utilized Docker for containerization, ensuring consistency across development and production environments, and implemented CI/CD pipelines for automated model deployment. Used Postman for API testing and validation• Collaborated with other business units to develop an Automatic License Plate Recognition (ALPR) system to enhance security measures and streamline regulatory compliance by leveraging Python, OpenCV, Pytorch, and YOLO for license plate detection and character recognition, ensuring high accuracy and real-time performance• Developed a machine learning algorithm using Python and Node.js to analyze employee attendance data in real-time, resulting in a 20% increase in accuracy compared to manual tracking methods.

India

As a key member of the MIDLabs Innovation Team, I led transformative projects leveraging Deep Learning and Computer Vision to revolutionize the hiring process:• Developed a Proof of Concept (POC) for a video analytics prototype that automated candidate evaluations. This solution significantly streamlined hiring processes, achieving a 40% reduction in manual review time and improving candidate assessment accuracy by 30%.• Engineered a functional prototype capable of analyzing candidate videos, extracting critical data such as facial expressions, body language, and speech patterns. This holistic approach provided a 25% increase in predictive accuracy for candidate fit.• Implemented a data-driven evaluation methodology, ensuring only the most qualified candidates advanced in the selection process. This approach led to a 35% improvement in hiring manager satisfaction and a 20% decrease in turnover rates.• Utilized Python, OpenCV, and PyTorch for data preprocessing and prototype development. Applied neural networks and natural language processing techniques to enhance candidate assessment.• Leveraged Docker for containerization to ensure consistency across development and production environments and utilized Kubernetes for efficient deployment management. Implemented CI/CD pipelines for automated testing and deployment, which reduced deployment times by 50% and increased system reliability.

Master'S Degree, Applied Computer Science

Coursework: Deep Learning, Applied AI, Machine Learning, Conversational AI, Computer Vision, Advance Programming Practices, Distributed.

Bachelor'S Degree, Computer Science