Sep 2023 – Dec 2023 · 3 mos · Michigan, United States · Remote

• Implemented advanced Excel and Power BI techniques to extract, consolidate, and standardize warranty field claim data from a complex network of 100+ OEM warranty systems.
• Optimized automated tracking system, reducing warranty costs and boosting customer satisfaction by 15%.
• Collaborate closely with the Corporate Warranty team, conducting thorough testing, including SAS analytics, to ensure accurate and reliable data for decision-making and quality assessment.
• Compiled a comprehensive report for Corporate Quality Warranty Synergy, utilizing Spark SQL for querying structured data and successfully reducing claims by 15% in 6 months.
• Analyzed key metrics into the report for informed decision-making on warranty trends in field claim data, reducing customer support issues by 20%.

Sep 2023 – Sep 2023 · 0 mo · Florida, United States · On-site

Jun 2023 – Aug 2023 · 2 mos · Remote

• Spearheaded the implementation of real-time computer vision algorithms for advanced object detection and tracking, ensuring precise identification of ad placements within multimedia content; boosted ad revenue by 25%.
• Applied advanced ResNet50 and CNN techniques on the AVA dataset, analyzing 255,500 images to enable precise facial recognition and eye tracking for targeted ad analysis, resulting in a 25% increase in ad engagement and conversion rate.
• Redesigned the storage infrastructure by leveraging Amazon S3 to enhance accessibility and integration of multimedia content and AVA data-set, resulting in 40% faster access and improved efficiency in model training using Amazon SageMaker.
• Revolutionized gaze detection by implementing LSTM and CNN architectures, achieving industry-leading accuracy rates of 70% and 85%, setting new standards for eye-tracking technology.

Jul 2021 – Aug 2021 · 1 mo · Delhi, India

• Collaborated in a 5-member team at the HIL R&D Centre on a Plant Recognition deep learning project, leveraging a
• data set of 30k images for robust training and testing.
• Achieved 89% accuracy in leaf health evaluation using auto-encoders and CNN, employing anomaly detection by assessing the
• reconstruction loss magnitude, ensuring precise anomaly identification . Created a flask based web application.

Oct 2020 – Nov 2020 · 1 mo · Delhi, India