Help & Supportcareers@gravityer.com
AI is analyzing your overall score…
Identifying your key strengths…
Evaluating your skill match against the job requirements…
Assessing your cultural and operational fit
Senior Principal Researcher at RBC Borealis
I received my PhD in Computer Science (machine learning and computer vision) from Simon Fraser University. Next I worked at Oracle Labs for 3 years (machine learning R&D). Now I am a machine learning researcher at Borealis AI, Vancouver. I have been mainly involved in the following research areas: ▬▬▬▬▬▬ Machine Learning: ✔ Deep Learning ✔ Graph Learning and Generation ✔ Explainable Machine Learning ✔ Probabilistic Graphical Models ✔ Latent Variable Models ✔ Kernel Learning ✔ Multi-Instance Learning ✔ Boosting ▬▬▬▬▬▬ Computer Vision: ✔ Video Event Detection ✔ Video Summarization ✔ Human Activity Recognition ✔ Image/Video Classification ▬▬▬▬▬▬ Finance: ✔ Risk and Capacity Models ✔ Credit Strategies ✔ Property Valuation ▬▬▬▬▬▬ Robotics: ✔ Robot Programming by Demonstration (a.k.a. Imitation Learning) ✔ Robot Motion Pattern Learning and Control ✔ Human-Robot Interaction ▬▬▬▬▬▬ Intrusion Detection: ✔ Network Traffic Analysis ✔ Log Data Analysis ✔ Time Series Analysis ✔ Anomaly Detection ▬▬▬▬▬▬ Natural Language Processing: ✔ Keyword Extraction ▬▬▬▬▬▬ Optimization: ✔ Convex Optimization ✔ Evolutionary Optimization ✔ Multi-objective Optimization ▬▬▬▬▬▬ Control Systems: ✔ Intelligent Control ✔ Optimal Control ✔ Cooperative Control ▬▬▬▬▬▬ Macro Economics: ✔ Electricity Market Analysis ✔ Game Theory
Simon Fraser University
Doctor of Philosophy (PhD), Computer Science
January 1, 2011 – January 1, 2015
University of Tehran
Master of Science (M.Sc.), Electrical and Computer Engineering
January 1, 2008 – January 1, 2010
University of Tehran
Bachelor of Science (B.Sc.), Electrical and Computer Engineering
January 1, 2004 – January 1, 2008
RBC Borealis
Senior Principal Researcher
April 1, 2025 – Present
RBC Borealis
Principal Researcher
January 1, 2021 – April 1, 2025
RBC Borealis
Senior Machine Learning Researcher
October 1, 2018 – January 1, 2021
Oracle Labs
Principal Member of Technical Staff / Machine Learning Researcher
January 1, 2018 – October 1, 2018
Vancouver, Canada Area
Oracle Labs
Senior Member of Technical Staff / Machine Learning Researcher
September 1, 2015 – January 1, 2018
Greater Vancouver Metropolitan Area
BroadbandTV
Part-time Research Engineer and Consultant
September 1, 2014 – August 1, 2015
Greater Vancouver Metropolitan Area
BroadbandTV
R&D Intern
May 1, 2014 – September 1, 2014
Greater Vancouver Metropolitan Area
Simon Fraser University
Teaching Assistant
January 1, 2012 – January 1, 2013
Simon Fraser University
PHD Student and Research Assistant
May 1, 2011 – September 1, 2015
IEEE Student Branch at University of Tehran
Lecturer
January 1, 2009 – January 1, 2010
Tehran
RAISE Institute at University of Tehran
Research Externship
January 1, 2007 – January 1, 2008
Tehran
University of Tehran
Teaching Assistant
January 1, 2006 – January 1, 2009
E-health Department at Iran Telecommunication Research Center
Research and Development Intern
January 1, 2005 – January 1, 2006
Tehran
Cyclist's Helmet Recognition Using Computer Vision
January 1, 2013 – Present
✔ Co-authored 1 paper. ✔ Using computer vision and machine learning algorithms in the application of detecting cyclists who are not wearing helmets.
Learning Probabilistic Latent Structured Models
January 1, 2013 – Present
✔ Designing novel algorithms based on Kernel trick and Gradient Boosting to learn latent structured models. ✔ A paper is under review.
Multi-Instance Classification by Learning Probabilistic Models
January 1, 2012 – Present
✔ Published 4 papers (in UAI, ICCV, CVPR, and T-PAMI). ✔ Applying multi-instance learning to video event detection, video summarization, image categorization, cyclist's helmet recognition, and human activity recognition from videos captured by street cameras. ✔ Designing a novel general framework for multi-instance learning based on probabilistic cardinality models, which can encode different cardinality assumptions, deal with diverse levels of labeling ambiguity, and be integrated with different machine learning algorithms (including max-margin and kernel learning). ✔ Designing a MIL algorithm based on boosting and linguistic aggregation functions, which can model intuitive and human-friendly assumptions in the data (e.g. ``some of'' the data instances are truly positive).
Statistical Machine Translation with Multiple Evaluation Criteria
January 1, 2012 – Present
Final project for the machine translation course. ✔ Proposing a method for discriminative training of translation systems with respect to multiple evaluation metrics.
TRECVid Multimedia Event Detection evaluation
January 1, 2012 – Present
✔ Co-authored 2 papers and 2 technical reports. ✔ Developing a system based on computer vision and machine learning tools to retrieve videos of interest from more than 100K videos of TRECVid project, sponsored by the National Institute of Standards and Technology. ✔ Collaborated with Genie team made up of research groups in Stanford University, Georgia Institute of Technology, SUNY-Buffalo and Honeywell led by Kitware Inc. ✔ Worked on 10Ex evaluation (i.e., learning only based on 10 training examples for each event) and video feature optimization, which resulted in about 25\% improvement in average precision for video retrieval.
Conceptual imitation Learning based on perceptual and functional characteristics of action
January 1, 2010 – Present
✔ Published 4 papers. ✔ Designing a bio-inspired conceptual model for robot learning by imitation, which can be used to teach high-level concepts (e.g., social skills) to robots based on perceptual or functional effects of actions. ✔ Developing a robotic system on the Aldebaran Robotics R Nao humanoid robot using Python and Matlab. ✔ Proposing a novel interactive algorithm to incrementally learn, abstract, and generalize spatio-temporal demonstrations of the teacher in the robot’s intelligent system. ✔ Incorporating the proposed model into different modalities, including vision, motor, and audition in order to make the robot learn and associate different perceptual representations of an action. ✔ Training the robot to learn emotional concepts based on the effects of its actions in a human facial expression.
Automatic Dynamic Coverage Control with Hierarchical Multi-Agent Reinforcement Learning
January 1, 2009 – Present
✔ Designing an intelligent system based on hierarchical multi-agent reinforcement learning for automatic and efficient (both in time and energy) sensing coverage of a set of target points with a set of sparse sensors which can be accessed by a distributed set of nearby artificial agents.
Learning Locally Linear Neuro-Fuzzy Models
January 1, 2009 – Present
✔ Designing an algorithm to train neuro-fuzzy models, which incrementally divides the data into clusters with affine input-output mappings and trains a fuzzy linear model for each cluster. ✔ Applying the proposed algorithm to predict the time series of sunspot solar activity used in solar physics and Evapotranspiration potential used in irrigation scheduling, which yielded state-of-the-art prediction results.
Multi-Criteria Group Decision Making
January 1, 2009 – Present
✔ Designing a novel technique for multi-criteria group decision making technique using fuzzy aggregators, which improves reflecting opinions of the majority of decision makers and provides more confidence for the final decisions.
Analysis of Electricity Markets, Using Evolutionary Algorithms
January 1, 2008 – Present
✔ Published 2 papers. ✔ Proposing novel algorithms to find Nash equilibrium in games with non-linear profit or demand functions. ✔ Proposing a multi-objective evolutionary algorithm to study the Pareto improvement model in an oligopolistic electricity market of nonlinear demand and an IEEE 30-bus power system with transmission constraints. ✔ Analyzing the IEEE 30-bus system with stochastic cost data in a risk management problem which maximizes the expected total profit but minimizes the profit risk in the market.
Automatic Artifact Identification in Image Communication
January 1, 2008 – Present
✔ Published 1 paper. ✔ Designing a system based on watermarking and machine learning to detect the artifacts (including Salt&Pepper, JPEG, Packet Loss, and AWGN) in image communication lines.
Cooperative State Estimation for Mobile Sensors with Optimal Path Planning
January 1, 2008 – Present
✔ Optimal control of a set of noisy mobile sensors for maximizing quality of estimation while minimizing the control effort.
Improving Ant Colony Optimization Algorithm
January 1, 2008 – Present
✔ Published 1 paper. ✔ Proposing an extended version of ACO with genetic semi-random restarts, which can escape from local optima, to solve MAX version of Multiplicative Squares problem. ✔ Combining different extensions of ACO to aggregate their advantages in a unified ant algorithm.
Invasive Weed Optimization for Intelligent Control and Decision Making
January 1, 2008 – Present
✔ Published 4 papers. ✔ Designing novel extensions of Invasive Weed Optimization (IWO) such as IWO/PSO, discrete IWO, co-evolutionary IWO, and multi-objective IWO. The algorithms achieved state-of-the-art optimization results. ✔ Using IWO/PSO algorithms in adaptive control of a surge tank. Simulation results showed the efficacy of the algorithm in terms of better accuracy and faster tracking. ✔ Using discrete IWO for cooperative multi-task assignment of drones. Simulation results showed that the algorithm outperforms genetics algorithms in both optimizing the objective function and computational time.
Probabilistic Graphical Models
Coursera
June 24, 2026 – Present
Cultural Fit Analysis
The candidate's background is heavily skewed towards academic research and advanced machine learning R&D. While this demonstrates deep technical expertise, the transition to a 'Data Analyst' role, which often requires more focus on business intelligence, reporting, and stakeholder communication, might present a cultural fit challenge. The projects are highly theoretical and research-oriented, with less emphasis on typical data analyst tasks like dashboard creation, A/B testing analysis, or SQL-based data manipulation. The breadth of skills is strong in ML/AI, but specific data analyst tools and methodologies are not explicitly highlighted.
Soft Skills & Operational Fit
The candidate's extensive research and publication history suggests strong analytical thinking, problem-solving, and independent work capabilities. Their experience as a Teaching Assistant indicates communication and mentorship skills. The collaborative nature of projects like TRECVid also points to teamwork abilities. However, the resume does not explicitly detail operational fit aspects such as project management, stakeholder communication, or agile methodologies, which are crucial for a senior role.