448 Ontario St, Toronto, ON
4372284989 | riddhi.mandal@mail.utoronto.ca linkedin.com/in/riddhi-mandal
Department of Chemical and Physical Sciences, University of Toronto
Recipient of multiple awards, and grants based on academic merit
Nominated for OSPP finals at EGU
Indian Institute of Science Education and Research, Kolkata, India
Conducted research at University of Cambridge
EGU Early career researcher grant
Recipient of DST INSPIRE Fellowship (Merit-based scholarship)
Indian Institute of Science Education and Research, Kolkata, India
Recipient of DST INSPIRE Fellowship (Merit-based scholarship)
Selected for Vijyoshi 2014 (exclusive scientific development camp)
DATech
Benchmarked ML model performance using Python, statistical analysis, and feature attribution to isolate quality bottlenecks
Identified optimization targets that improved accuracy and increased Net Promoter Score
OutlierAI
Optimized physics-informed ML models using feature engineering and model tuning
Improved training efficiency and accuracy, accelerating deployment in resource exploration and signal analysis
University of Toronto
Developed real-time anomaly detection for time-series using acoustic emission data and deep learning
Applied wavelet transforms and neural networks for robust signal identification under high noise
Applications include early-warning systems, structural failure prediction, and anomaly tracking in complex environments
University of Toronto
Constructed a high-dimensional simulation environment modeling stress evolution in complex 3D systems
Combined FEM, DEM & FDM solvers, and sensitivity analysis to forecast system instabilities under uncertain inputs
Applied stress interaction and kinematic modeling tools for analysis under varying parameters
University of Toronto
Simulated material response under stress using 5,000+ variable models with dynamic friction and stress transfer
Investigated stability domains using parametric stress testing, matrix sensitivity scans, and data-driven thresholds
Delivered insights for risk mitigation in subsurface operations and structurally sensitive environments
University of Toronto
Engineered a high-accuracy peak detection algorithm handling >2000% frequency variance across signal types
Achieved 96% accuracy and 92% recall, enabling robust detection across time series such as seismic, financial, and biomedical
Outperformed established peak detection techniques, being almost two orders of magnitude faster than most techniques
University of Toronto
Conducted electromagnetic and gravity surveys, producing subsurface models with interpreted structures
Implemented geospatial data pipelines and streamlined modeling tools for exploratory field operations
Conducted Monte Carlo sweeps and grid-based sensitivity analysis to define failure thresholds
Indian Institute of Science Education and Research, Kolkata
Conducted integrated surveys using seismic, gravity, and InSAR datasets to model regional stress accumulation
Performed stress tensor and moment tensor inversions to reconstruct stress fields
Modeled rupture propagation of a large event using wavelet transforms and back-projection to reveal energy release dynamics
Provided critical insights into stress transfer, improving regional hazard assessment and advancing deformation models.
D. Powali, S. Sharma, R. Mandal, S. Mitra, “A Reappraisal of the 2005 Kashmir (Mw 7.6) Earthquake and its Aftershocks: Seismotectonics of NW Himalaya”, Tectonophysics
S. Dey, D. Powali, J. Chaudhury, M. Ghosh, R. Mandal, J. Kanaujia, S. Mitra, “28 August 2018 (Mw 4.5) Bengal Basin earthquake highlights active basement fault beneath the sediments”, Current Science, 2019, 116,1633-1636