Jasmin Chaughule
Data Scientist @Dozee
Signup · Get unlimited contacts
WORK HISTORY
Data Scientist @Dozee
Bengaluru, IN
Engineered spectral,frequency, and wavelet based features from 15 second segments of Ballistocardiogram (BCG) signals to asses signal quality in order to ensure accurate generation of patient’s vitals.• Optimized workflow for segment extraction and feature computation which reduced vitals’ experiment turnaround time by 50%.• Leveraged data visualization tools like Matplotlib and Seaborn to plot histograms, box plots, violin plots, and heatmaps to explore feature distributions, assess feature importance, and determine optimal thresholds.• Implemented supervised learning models like SVMs and xgboost on extracted features to assess signal quality, contributing to a 12% improvement in vital detection rates in production.• Designed auto-encoders and transformers using PyTorch for anomaly detection to identify poor quality or abnormal signals.• Created data preprocessing, cleaning and validation pipelines that enabled the curation of more than BCG segments
EDUCATION
BITS Pilani, Hyderabad Campus
Bachelor of Engineering - BE
ABOUT JASMIN CHAUGHULE
As a Data Scientist, I believe in developing innovative solutions by coupling data-driven insights with the power of machine learning models and artificial intelligence. I am a graduate of BITS Pilani, which equipped me with the technical knowledge and problem-solving skills required for this field.My experience as an intern working with LLMs,E mbeddings, Vector Databases and Retrieval-Augmented Generation for an AI copilot designed for private market investors gave me valuable insight into the workings of AI chatbots — an increasingly powerful tool in today’s world.Currently, I’m working as a Data Scientist at Dozee, a remote patient monitoring startup, where I work with Ballistocardiogram (BCG) signals to extract patient vitals. This role has given me the opportunity to closely witness how machine learning can empower critical fields like healthcare, while strengthening my skills in signal processing, anomaly detection, machine learning and deep learning.
This profile is compiled from publicly available professional sources. Unifers is not affiliated with or endorsed by LinkedIn. Request removal of this profile.