Help & Supportcareers@gravityer.com
Accelerating AI at d-Matrix, ex-Intel
AI is analyzing your overall score…
Identifying your key strengths…
Evaluating your skill match against the job requirements…
Assessing your cultural and operational fit
Fields which interests me the most are Computer Vision, Machine Learning, Natural Language Processing, Image Processing Embedded Systems, Optimization of signal processing algorithms for efficient hardware implementation. Through my knowledge and skills, I want to contribute to the technological advancements in the innovative projects undertaken by companies.
University of Michigan
Master’s Degree, Electrical and Computer Engineering
January 1, 2015 – January 1, 2017
K.J. Somaiya College of Engineering
Bachelor’s Degree, Electronics and Telecommunication Engineering
January 1, 2011 – January 1, 2015
K.J. Somaiya College of Science and Commerce
High School, Science
January 1, 2009 – January 1, 2011
d-Matrix
Sr Staff Software Engineer - ML/AI Workloads
July 1, 2023 – Present
Intel Corporation
AI Software Architect
April 1, 2018 – July 1, 2023
Intel Corporation
Machine Learning Engineer
July 1, 2017 – April 1, 2018
University of Michigan
Graduate Student Research Assistant
January 1, 2017 – April 1, 2017
Ann Arbor, Michigan
University of Michigan
Research Assistant
September 1, 2016 – December 1, 2016
Ann Arbor, Michigan
Intel Corporation
Graduate Technical Intern
May 1, 2016 – August 1, 2016
Bengaluru Area, India
University of Michigan
Graduate Student
August 1, 2015 – April 1, 2017
Ann Arbor, Michigan
Eduvance
Backend Digital VLSI Design
February 1, 2015 – April 1, 2015
Mumbai Area, India
Eduvance
Summer Industrial Training in Embedded Systems
June 1, 2014 – July 1, 2014
Mumbai Area, India
Research Innovation Incubation Design Lab (RiiDL)
Project Innovator
December 1, 2013 – January 1, 2014
Mumbai Area, India
Visual Question Answering
February 1, 2017 – April 1, 2017
- The VQA task seeks to solve the problem of automatically generating answers to questions of images – an important problem in realizing Artificial Intelligence. - Images were fed through a CNN, questions and answers through a RNN to a : - Conditional GAN setting - Co-Attention setting - Used tensorflow for the implementation
Parallel Implementation of Adaboost Classifier on Distributed, Shared and GPU systems
September 1, 2016 – December 1, 2016
- Parallelized the serial ADABOOST classification algorithm to be implemented on three platforms(Distributed memory, Shared memory and GPU). - Evaluated performance, scaling and speedup on all the three systems. - Analyzed which system would give best performance when the data size and number of weak classifiers are changed
Classification of the MNIST dataset using Convolutional Neural Network
March 1, 2016 – Present
- Designed a convolutional neural network architecture which trains on the MNIST train data and obtained accuracy of 97.74% on the test set
High Resolution Images from multiple low resolution Images using Delaunay Triangulation
February 1, 2016 – Present
- The images taken by the world's smallest computer 'The Michigan Micro Mote' have low resolution and are also noisy. - Implemented the technique of Delaunay triangulation to obtain a single high resolution image from multiple low resolution images and applied deblurring on the high resolution image to nullify blurring artifacts. - Images were obtained from multiple millimeter-sized imagers displaced by a fixed distance.
Motion Segmentation using Generalized Principle Component Analysis
November 1, 2015 – Present
- Implemented video Motion Segmentation using GPCA technique for all possible affine motions in MATLAB. - Studied the concepts of power factorization, dimensionality reduction and implemented them in a unified framework. - Obtained accurate segmentation results with misclassification error rate around 9% for different input videos.
Feature Matching and Foreground and Background Separation
October 1, 2015 – Present
- Designed a game 'SPOT-IT' by implementing Feature Matching and Foreground separation in MATLAB. - Implemented Feature matching using Harris Operator and Histogram of Gradient Descriptors and Foreground and Background separation using SLIC (Simple Linear Iterative Clustering) and max-flow-min-cut graphing technique.
Implementation of Adaptive Filters using LMS Algorithm on FPGA
March 1, 2015 – Present
- Implemented Adaptive Filter based on Least Mean Squared Algorithm on Field Programmable Gate Arrays (SPARTAN 3) using Verilog. - Used Linear LMS algorithm to minimize hardware requirement but got optimum results for noise with different nature and SNR when compared to results from MATLAB. - Adaptive Filter trains its coefficient iteratively to reduce the mean squared error between the input and desired signal and gives optimum filter coefficients after the training period. Hence every time nature of signal and noise changes, designing a new filter is not necessary.
Smart Seating Arrangement System
July 1, 2014 – Present
- Designed a system which displays real time information of occupied and empty seats in an Auditorium. - Used Push Button Array to communicate data to MATLAB via AURUM Board (A micro-controller board designed by WEL Labs IIT Bombay based on PIC18F4550 ) - Auditoriums are dark and if crowded with only a few seats empty then it becomes difficult to find empty seats. Hence, an automatic empty seat indicator which displays this information outside the auditorium would be really helpful
Proximity Sensor using PSOC 4 and SIM 300 GSM module
June 1, 2014 – Present
- Designed a theft detection and wireless notification system using cypress PSOC 4 board (CapSense Technology was used) and SIM 300 GSM module. - Modified the project by changing the threshold as a water sensor.
Response time meter
October 1, 2013 – Present
- Designed a small prototype to detect response time of an individual based on time difference calculation of trigger and response - Used Monostable Multivibrator, High Pass Filter and Variable Resistor. - By setting resistance of variable resistor to different values, time limit for responses for people from different age group can be set.
Wireless Time Division Multiplexing System
August 1, 2013 – Present
- Designed a Wireless Time Division Multiplexing System with variable time slots, to overcome interference caused by multiple transmission, single reception at a single frequency(434MHz). - Used low cost RF module (434MHz) and ARDUINO Micro-Controller Board. - Frequency spectrum is limited, hence to communicate information from different sensors to a single destination using different frequency channels would be unnecessary wastage of the spectrum. - Also, the transceivers available to solve the problem using Time Division Multiplexing are costly. Hence, design of such low-cost TDM system helps solve the stated problem
Converged Communication
Centre for Excellence in Telecom Technology and Management
June 24, 2026 – Present
Backend Digital VLSI Design
Eduvance
June 24, 2026 – Present
Transformers: Text Classification for NLP Using BERT
June 24, 2026 – Present
Modern Digital Design and Embedded Systems
e-Prayog, Wadhwani Electronics Lab, IIT Bombay
June 24, 2026 – Present
Summer Industrial Training in Embedded Systems
Eduvance
June 24, 2026 – Present
Cultural Fit Analysis
The candidate's project portfolio is highly diverse, ranging from embedded systems and hardware design to advanced machine learning and computer vision. This breadth of interest and capability suggests an adaptable individual who can contribute to various technical challenges. The experience at Intel, a large corporation, indicates an ability to work within established frameworks. However, the target role is 'Data Analyst', while the candidate's experience leans heavily towards ML/AI Engineering and Architecture. While there's overlap in data handling and analysis, the core focus of their past roles and projects is more on model development, system design, and hardware acceleration rather than pure data analysis, reporting, and business intelligence. This might indicate a potential mismatch in the day-to-day responsibilities and expectations of a typical Data Analyst role, which often requires strong business acumen and data visualization skills not explicitly highlighted.
Soft Skills & Operational Fit
The candidate's project descriptions indicate a problem-solving mindset and an ability to identify and address real-world challenges (e.g., low-cost TDM system, smart seating arrangement). Experience at Intel suggests an ability to work within a corporate structure and align with strategic goals. The description of 'ramping up quickly' and 'delivering crucial code' points to adaptability and a results-oriented approach. However, without direct assessment data, specific soft skills like teamwork, leadership, or stress handling cannot be definitively evaluated.