Sayan Mitra

- Leveraged machine learning models to improve sales forecasting accuracy for the Finance and Operations (F&O) department. This data-driven approach empowered informed decision-making, leading to significant business growth.- Developed an automated solution for identifying adverse events in audio calls, eliminating the need for manual review and achieving a 100% reduction in effort and easier readability of calls. This improved the response time of events and allowed LoB to save dollars on the non identific

Digital Transformation and Innovation in shaping the future of Compliance. Dealing with topics which involve creating/supporting tools like G&E and data visualisation through dashboard reporting.

- Designed and built high-performance, event-driven data pipelines for a leading Dutch e-commerce giant. This real-time data architecture enabled an impressive 83% improvement in new product delivery times, significantly enhancing customer satisfaction.- Built and deployed a comprehensive solution for a P2P used car marketplace, streamlining communication between buyers and sellers. This innovative platform resulted in a remarkable 10x decrease in lead response time, boosting conversion rates and user exper

Data Analysis on sensors. Time series forecasting and fault analysis in order to ensure smooth movement and transparency in logistics.

Metric Evaluation and KPI identification of ML models. Ground Truth detection for object/action recognition in targeted advertising.

Data Analysis and Query Optimisation in Teradata DB. Providing automation in Apple Care Team for data manipulation and cleaning tasks using Unix/Python scripting.

- Deep Learning for Computer Vision - Object Detection, Tracking and Segmentation - Currently writing master's thesis on the topic of "A Novel Fusion Approach to Enhance the Emotion Prediction Accuracy in Thermal/RGB Imaging" at the chair of Robotics, Artificial Intelligence and Real-time Systems.