Aurangzeb Ansari

Senior Machine Learning Engineer, Bangalore, KA
Philips Healthcare, November 2018-Present

• Led a team of 4 machine learning engineers and developed a heart disease prediction system comprising two deep learning neural network architectures: CardioNetX and eCardioNetX.
• Designed and developed INSIGHT CXR, a machine learning architecture for detecting abnormalities in chest radiographs.
• Worked on a predictive maintenance model for X-Ray machines, reducing downtime by 20%.
• Designed and implemented a deep learning algorithm for real-time image enhancement, improving diagnostic accuracy by 15%.
• Utilized AWS SageMaker for managing deep learning iterations and model training.
• Coordinated data collection, preprocessing, and augmentation efforts.
• Optimized existing models to improve accuracy and reduce computational time.
• Developed metrics to evaluate model performance and conducted extensive testing and validation.
• Integrated machine learning models into production systems using Kubernetes and Docker.
• Developed a windows service for uploading medical data to the cloud from hospitals.
• Built web applications for real-time predictions with trained models.
• Trained new developers on company standards for design and review.

SENIOR SOFTWARE ENGINEER, MYSORE, KA
L&T Technology Services, January 2016-October 2018

• Developed a predictive maintenance system for vehicles using machine learning.
• Analyzed sensor data from various vehicle components to identify patterns and predict potential failures.
• Built software for real-time communication between sensors in cars, assisting in early fault detection.
• Collaborated with project managers, developers, quality assurance, and customers to resolve technical issues.

SOFTWARE DEVELOPER, KOLKATA, WB
Wipro Technologies, January 2013-October 2016

• Revised, modularized, and updated old code bases to modern standards.
• Developed desktop applications in the shipping domain for ABS, a ship certification company.
• Wrote Unit tests and removed code duplication, improving code quality.