Shopify | Investor | Realtor | Ex-Microsoft, Yahoo, StumbleUpon
AI is analyzing your overall score…
Identifying your key strengths…
Evaluating your skill match against the job requirements…
Assessing your cultural and operational fit
Book Time with Me: https://calendly.com/asonix022
University of Minnesota
Ph.D., Electrical Engineering
January 1, 2011 – January 1, 2015
University of Minnesota
MS, Electrical Engineering
January 1, 2010 – January 1, 2011
University of Minnesota
Masters Minor, Mathematics
January 1, 2010 – January 1, 2011
DA-IICT
B-Tech, Information and Communication Technology
January 1, 2006 – January 1, 2010
Shopify
Director of Engineering
May 1, 2026 – Present
Shopify
AI Leader
April 1, 2025 – May 1, 2026
Microsoft
Principal Applied Science Manager
March 1, 2022 – August 1, 2024
Microsoft
Applied Science Manager
December 1, 2021 – March 1, 2022
Microsoft
Senior Machine Learning Scientist
August 1, 2018 – December 1, 2021
Yahoo
Research Scientist - Content Recommendation/ML
December 1, 2015 – August 1, 2018
United States
StumbleUpon
Research Engineer
March 1, 2015 – December 1, 2015
San Francisco Bay Area
University of Minnesota
Research Assistant
September 1, 2014 – May 1, 2015
Minneapolis
StumbleUpon
Data Science Intern
June 1, 2014 – August 1, 2014
San Francisco Bay Area
University of Minnesota
Research Assistant
September 1, 2012 – May 1, 2014
Mitsubishi Electric Research Laboratories
Summer Intern
June 1, 2012 – September 1, 2012
Greater Boston Area
University of Minnesota
Research Assistant
October 1, 2010 – May 1, 2012
DA-IICT
Teaching Assistant: Digital Logic Design
December 1, 2009 – May 1, 2010
Gandhinagar, Gujarat, India
DA-IICT
Teaching Assistant: Basic Electronic Circuits
August 1, 2009 – December 1, 2009
Gandhinagar, Gujarat, India
Geometric Wavelets: concepts and applications
January 1, 2012 – May 1, 2012
Multiscale geometric wavelet is a novel method for analyzing high-dimensional point clouds using a low-dimensional manifold. In this approach, a structured multiscale dictionary (columns of which are the geometric wavelets) is learned on training data which can be very e?ciently computed. The expansion of a data point in this learned dictionary guarantees a certain degree of sparsity which is fundamental to current research trends in the ?eld of machine learning, computer vision and compressed sensing. In this project, we wish to analyze the conceptual framework of these geometric wavelets.
Level Set Estimation Using Total Variation Norm Regularization
August 1, 2011 – December 1, 2011
Level set estimation is an important problem with applications in the field of medical imaging, astronomy and digital elevation maps. In many scenarios, the direct and complete measurements of the signal are not available and the measurements come in the form of pro- jections, for instance in tomography and compressed sensing. This paper describes a simple procedure to estimate the level set of a sig- nal with highly incomplete measurements in the presence of addi- tive noise. The proposed procedure is based on Total Variation (TV) norm regularization and directly gives level set without the interme- diate step of reconstruction.
Compact Wideband Bandpass Filter Using Stepped Impedance Resonators and Interdigital Coupling Structures: RF and Microwave devices
January 1, 2011 – August 1, 2011
Modelled the interdigital bandpass ?lter layout in ADS Momentum and Sonnet. A semester long project with a project report and video presentation.
IEEE 802.22 WRAN Standard: Wireless Communications
January 1, 2011 – May 1, 2011
Studied the physical layer design and challenges of the very new wireless standard in a group of two. A presentation and project report was part of the project.
Channel Equalization Schemes: Digital Communications
August 1, 2010 – December 1, 2010
Simulated different channel equalization schemes in Matlab and did a comparative study.
LDPC Codes: Information Theory and Coding
August 1, 2010 – December 1, 2010
Studied LDPC codes in detail for semester long class project. A project report was require
B. Tech Research Project
August 1, 2009 – May 1, 2010
In this report, adaptive channel estimation algorithms for OFDM systems have been proposed. First, we implemented IQR-2D-RLS algorithm which has similar BER performance and computational complexity (O(N2)) as 2DRLS algorithm but is numerically more stable than the latter. IQR-2D-RLS propagates the square-root of inverse of input autocorrelation matrix which preserves the property of positive definiteness or Hermitian symmetry unlike 2D-RLS algoithm. Then, we tried to improve the computational complexity of the adaptive channel estimation algorithms. For this, we proposed Set Membership Feasibility estimation algorithms, namely 2DSM-NLMS and 2D-SM-AP. These algorithms compromise on the convergence speed which is usually not required in a stationary environment but reduces the computational complexity to O(N) i.e. an order less than the conventional 2D-RLS algorithm. Here, the adaptive channel coefficients are updated only when the estimated error is greater than a predefined threshold. Also, we tried to analyze the performance of MIMO two-way relaying in a frequency selective environment. Spatial filtering was utilized at the relay station (RS), which was implemented using MMSE and ZF (Zero Forcing) criteria. Here, CSI (channel state information) is required only at the RS and not on the transmitting and receiving nodes. It was observed that the performance is improved using OFDM in a frequency selective environment.
UART: Digital System Architecture
August 1, 2008 – December 1, 2008
In a team size of 2 implemented UART on FPGA kit using Verilog and synthesis tool Xilinx & ModelSim.
MOTAP: Software Engineering
August 1, 2008 – December 1, 2008
Developed, Mobile and web based system for online trading of agricultural products. The main aim of the project was to remove the middleman from the chain. It was a 5 member team project and was successfully completed with lot of appriciation.
Diabetes Health Care System (DHCS)
January 1, 2007 – May 1, 2007
Developed an online health care system for a widely spreading disease Diabetes. The system will help the patients to communicate with their doctors, hospitals and labs; get their medical details stored with the system and keep a check on their progress report.
edX Verified Certificate for Introduction to Big Data with Apache Spark
edX
June 24, 2026 – Present
Cultural Fit Analysis
The candidate's extensive experience in research and development, coupled with leadership roles in AI/ML, indicates a strong drive for innovation and technical excellence. The academic projects show a diverse range of interests in signal processing, machine learning, and digital systems. While the target role is 'Data Analyst', the candidate's experience leans heavily towards advanced Machine Learning, Applied Science, and Engineering Leadership. This might indicate an overqualification for a standard Data Analyst role or a potential mismatch if the role does not involve significant advanced modeling or research. The project descriptions lack explicit collaboration details beyond 'team size of 2' or 'group of two', making it difficult to fully assess collaborative cultural fit. The certifications include 'Introduction to Big Data with Apache Spark', which is relevant for data roles.
Soft Skills & Operational Fit
The candidate's career progression from Research Scientist to Director of Engineering and AI Leader suggests strong leadership, strategic thinking, and project management capabilities. The academic background and research roles imply strong problem-solving, critical thinking, and independent work skills. The project descriptions, while technical, are generally clear, indicating good written communication. However, without direct assessment data on soft skills or operational fit in a team setting, a comprehensive evaluation is limited.