Debasmita Pal
@debasmitapal
I build lightweight AI models for real-time cardiac monitoring on resource-constrained medical devices.
IndiaWhat I'm looking for
I'm building lightweight AI and machine learning systems for cardiac monitoring, from clinical ECG/PPG data collection to real-time deployment on resource-constrained medical devices.
At the University of Calcutta and CSIR, I developed embedded arrhythmia detection models reaching 99.21% accuracy with 782ms latency and 390kB memory on ARMv6 hardware. I also developed a myocardial infarction detection and localization model achieving 99.74% accuracy for real-time ECG analysis.
My research includes improving arrhythmia minority-class F1-score by 11% through hybrid data augmentation, alongside peer-reviewed IEEE publications in cardiac signal processing and medical-device machine learning.
At TCS Research, I built a PyTorch and OpenCV pipeline for sports and yoga motion-kinetics estimation using OpenSim biomechanical modelling. I'm pursuing a part-time PhD focused on automatic detection of abnormal cardiac episodes using lightweight signal processing and machine learning.
Experience
Work history, roles, and key accomplishments
Research Fellow
University of Calcutta
Jun 2026 - Present (3 months)
Built low-resource multi-class arrhythmia detection using machine learning and deep learning techniques.
Built a PyTorch/OpenCV deep learning-aided pipeline for motion-kinetics estimation from sports/yoga video, using OpenSim for biomechanical performance modelling.
Senior Research Fellow
CSIR – Council of Scientific & Industrial Research
Jun 2024 - Feb 2026 (1 year 8 months)
Developed lightweight, real-time arrhythmia detection models for embedded platforms, reaching up to 99.21% accuracy at sub-second latency. Designed ML and signal-processing pipelines optimized for ARMv6 microcontrollers for wearable cardiac monitoring devices.
Junior Research Fellow
DST & Biotechnology, West Bengal
Apr 2024 - May 2024 (1 month)
Developed ML-based methods for early arrhythmia detection and ECG signal acquisition. Built prototype pipelines enabling low-latency, on-device inference.
Project Assistant
Regional Geriatric Centre, Medical College, Kolkata
Dec 2022 - Dec 2023 (1 year)
Collected ECG/PPG data at Medical College, Kolkata using research-grade biomedical signal acquisition instrumentation, for arrhythmia detection analysis in the Indian population.
Education
Degrees, certifications, and relevant coursework
University of Calcutta
Doctor of Philosophy, Cardiac Signal Processing
2023 -
Pursuing a part-time PhD in automatic detection of abnormal cardiac episodes using lightweight signal processing and machine learning techniques.
University of Calcutta
Master of Technology, VLSI Design
2020 - 2022
Grade: DGPA 8.90
Completed M.Tech in VLSI Design with a DGPA of 8.90, ranking first in the program.
Maulana Abul Kalam Azad University of Technology, West Bengal
Bachelor of Technology, Electronics and Communication Engineering
2016 - 2020
Grade: DGPA 8.92
Completed B.Tech in Electronics and Communication Engineering with a DGPA of 8.92.
Tech stack
Software and tools used professionally
Availability
Location
IndiaAuthorized to work in
Salary expectations
Job categories
Interested in hiring Debasmita?
You can contact Debasmita and 90k+ other talented remote workers on Himalayas.
Message DebasmitaGet matched with your dream remote job
Sign up now and join over 250,000+ remote workers who receive personalized job alerts, curated job matches, and more for free!
