Pranjool Gupta

Lenexa medical - Real-time Monitoring of patient to detect pressure injury using Machine Learning/Deep Learning: Developed ML models (Boosting, Logistic Regression, Random Forest, SVM) for patient risk assessment. Designed custom object classification architecture (VGG/ResNet style) and trained object detection models (Nanodet, YOLO-V5) for human posture/body part localization. Implemented Redis Pub/Sub and designed an end-to-end architecture on AWS, leveraging services such as Lambda, Kinesis Stream and SageMaker. Vigo Care - ECG Analysis: Developed 1-D and 2-D model for noise detection from ECG Signals, and is currently spearheading the development of a robust model for accurate arrhythmia detection.

Tata capital - Call Analytics: Spearheaded the development of sophisticated machine learning models to accurately classify customer language and gender. Employed librosa for precise extraction of MFCC features. Orchestrated seamless audio-to-text transcription through custom Python script and cloud solutions like AWS and Azure. Leveraged Natural Language Processing techniques to quantitatively assess spoken words and performed sentiment analysis on customer interactions.