Experience

PiinPoint

• Canada
• Software Development
• 1 - 100 Employee

• Data Scientist Lead

  • Nov 2022 - Present

  - Lead building ML solutions using predictive/forecasting models for PiinPoint enterprise customer's problem and retailers (market optimization, sales prediction, location recommendation for market expansion using GIS data/client data)



Mappedin

• Canada
• Software Development
• 1 - 100 Employee

• Senior Machine Learning Engineer

  • Oct 2021 - Nov 2022

  •Build, test, train and deploy various CNN based deep learning frameworks for indoor mapping solutions (semantic segmentation, object detection, raster to vector data conversion, image enhancement, feature extraction, data augmentation) •Build a ML pipeline for creating digitized map floor plan using Generative Adverbial Network (GAN) and NLP frameworks

• Machine Learning Developer

  • Oct 2020 - Oct 2021



Cerebri AI Inc.

• United States
• Software Development
• 1 - 100 Employee

• Data Scientist

  • Jan 2019 - Apr 2020

  Toronto, Canada Area Worked with various predictive models and time series analysis models (Random Forest, XGBoost, CNN , Arima) for forecasting of macroeconomic indicators. Designed and developed predictive models using gradient boosting algorithms for customer performance assessment (lifetime value prediction, churn prediction, propensity prediction) Actively involved in data preparation, creating customer journey, data cleaning and feature engineering pipelines Worked on cutting edge… Show more Worked with various predictive models and time series analysis models (Random Forest, XGBoost, CNN , Arima) for forecasting of macroeconomic indicators. Designed and developed predictive models using gradient boosting algorithms for customer performance assessment (lifetime value prediction, churn prediction, propensity prediction) Actively involved in data preparation, creating customer journey, data cleaning and feature engineering pipelines Worked on cutting edge techniques (Shapely value) for AI model explainability Collaborated on building a Next Best Action model using both supervised method (cross entropy) and deep reinforcement learning method (DQN) Designed and developed clustering methods with PCA for customer segmentation to analyze marketing data Show less



Miovision

• Canada
• Software Development
• 200 - 300 Employee

• Computer Vision | Machine Learning Researcher

  • May 2018 - Jan 2019

  Designed and deployed CNN models for vehicle localization, classification and tracking using camera video imaging Primary concepts: deep learning, data augmentation, object detection, object tracking, /Python/C++/TensorFlow



University of Waterloo

• Canada
• Higher Education
• 700 & Above Employee

• Machine Learning Researcher

  • May 2017 - May 2018

  Schlegel-UW Research Institute for Aging and Vision and Image Processing Lab -Developed emotion detection using deep learning frameworks such as LSTM and OpenPose Library -Designed, developed and tested machine learning and data mining methods for analyzing wireless sensor data and wearable sensors, gait speed analysis using thermal imaging and radar data Primary concepts: machine learning, data fusion, feature extraction, filtering, data mining, Python/C++/Matlab



Airo Health

• Data Scientist

  • Jan 2017 - May 2017

  Kitchener, Canada Area -Designed, developed and tested different machine learning methods for an automatic PPG-based calorie tracker device Primary concepts: feature Extraction, data classification, Python, Matlab



University of Waterloo

• Doctoral Researcher

  • Sep 2012 - Dec 2016

  Canada -Designed and developed a Bayesian reconstruction framework for image enhancement using conditional random field model. -Collaborated in design and development of different image processing algorithms including image classification, super resolution imaging, image enhancement, image denoising and image reconstruction -Wrote multiple technical reports which were published in academic journals and conference proceedings



Christie Digital Systems

• United States
• Appliances, Electrical, and Electronics Manufacturing
• 700 & Above Employee

• Research Intern

  • May 2015 - Dec 2015

  Kitchener, Canada Area -Persuading research on 3-D geometrical reconstruction using projector-camera systems. -Designed and developed a saliency-guided projection geometric correction using a projector-camera system to compensate for geometrical distortion. Primary concepts: camera development, image processing, computer vision, C++, OpenCV and Point Cloud analysis



University of Manitoba

• Canada
• Higher Education
• 700 & Above Employee

• Research Assistant

  • Jan 2012 - Sep 2012

  Electrical and Computer Engineering Department Designed and developed a depth-dependent matched filter for SNR improvement in optical coherence tomography imaging