Nitish Sandhu

I am currently working at a startup specializing in Computer Vision using Deep Learning.

My contributions include:

1. Deep Learning Model Training and Optimization: Led the training and fine-tuning of advanced deep learning models to significantly enhance capabilities in image selection, image editing, and image sharing, resulting in marked improvements in model accuracy and overall application functionality

• Advanced Training Techniques: Employed cutting-edge methodologies and algorithms to train models, ensuring they reached peak performance for complex image-related tasks
• Enhanced Application Functionality: The improved models contributed to superior application features, providing users with more accurate image selection, seamless editing options, and efficient sharing capabilities
• Performance Metrics: Achieved significant improvements in key performance metrics, demonstrating the effectiveness of the training and optimization processes

2. Backend Development and Architecture: Spearheaded the design and implementation of the Python Backend Desktop Applications for Image Culling, Editing and Share, from inception, ensuring robust performance, scalability, and maintainability across three OS namely as Windows, Silicon Mac and Intel Mac. Desktop Applications available at https://algomage.com/downloads.

This involved:

• Architecting a comprehensive and modular backend structure, leveraging best practices in software engineering to have seamless integration with front-end components and thirdparty services
• Optimizing performance through efficient code, smaller and faster OS specific inference deep learning models, data management, and system resources.
• Implementing rigorous testing and debugging protocols to guarantee high reliability and stability and security measures to protect code and models.

• Worked as a Research Assistant in Video Analytics Lab, CDS Department, Indian Institute of Science, focusing on Neural Radiance Field (NeRF).

• Worked in an early stage startup in the area of Computer Vision
• Worked on different areas of image recognition and vision based navigation.

• Worked in RPM Prediction team, focusing on Behavioural Systems which deploys ML based models that learn about website user from its history on various metrics such as revenue, cost, impressions and bids accordingly during real-time bidding in online advertisement.