Associate Vice President

• Apr 2021 - Present

Data Science Manager

• Dec 2019 - Apr 2021

Extraction of Syntactic and Semantic search queries. Machine learning models for query correction, products search, products ranking, user profiling, and revenue impact. Architecture for data flow, annotation, modelling, and deployment flow.

Co-Founder

• Aug 2016 - Nov 2019

There are some of the project which I did at mBreath.Real-Time Human Vital Parameter Estimation Using Pressure Sensor and Machine LearningMeasured respiration and heart rate through an array of pressure sensors and Deep Learning (CNN, RNN-LSTM) algorithms. [Filed Patent Ref No. 201731038573]. 96% accuracy was achieved. The project had three stages 1) preprocessing with analog filters, 2) Digital Signal Processing, 3) Machine Learning model thereafter data is uploaded to the cloud for further processing.Real-time Classification of Environmental Sounds & Their Impacts on Sleep QualityCollected audio using a microphone and Qualcomm’s Snapdragon processor enabled custom-built board running Android 6.0. Audio Cleaning, Feature Extraction (more than 700 features) in Temporal, Spectral, Energy, Harmonic, and Cepstral domain, Feature Selection for faster training and better accuracy and ML models (CNN, LSTM, RF, SVM) are designed to classify snoring and other sound and their impact on sleep quality.A Real-Time System and Methods for the Detection of Sleep Apnea and Sleep StageUsed CNN, Inception, LSTM, CNN-LSTM models to estimate Sleep Apnea and Sleep Stages using the respiration signal only. The training data was obtained from MESA, National Sleep Research Resource, USA. I did data cleaning and balancing thereafter feature engineering and selection were done before applying the ML model. Achieved 82.7% accuracy in Sleep Apnea which is higher than any known methods applied on respiration signal in MESA Dataset and achieved more than 80% accuracy in sleep stage classification which can be considered as good accuracy.Visit:http://www.sleepdoc.aihttp://www.facebook.com/mBreathOfficialhttp://www.twitter.com/mBreathOfficial

Chief Executive Officer

• Aug 2016 - Nov 2019

I was responsible for setting the overall company policy.

Data Scientist

• Aug 2016 - Nov 2019

Developed two end-to-end products "Smart Wireless System and Method for Human Health Monitoring Using Doppler Radar“ [Filed Patent Ref. No. 201731038573] and "Smart Sleep Sense System For Human Health Monitoring And Screening Using Velostat Pressure Sensor" [Filed Patent Ref. 01731041913 ]Worked on multiple fronts including Business Model Development, Customer Acquisition, Market Analysis, Product Feasibility Analysis, and Customer Segmentation and Satisfaction analysis while working as Data ScientistStudied Socio and Economic Impact of healthcare and Sleep Disorder in the US market Did cost optimization while designing the product SleepDoc by keeping time-to-market in mindHaving experience in working with GitHub and Bitbucket for code management.Used Deep Learning (CNN, LSTM, CNN-LSTM, Inception, VGGs), Machine Learning models (Logistic Regression, Random Forest, SVM, and XGBoost), and Feature Selection for real-time audio analysis and classification, sleep apnea detection, sleep stage detection, snoring detection.Developed SleepDoc, SleepDoc uses the FMCW RADAR system to measure respiration and heart rate. Designed Android Application and state-of-the-art ML Algorithms (SVM, RF, CNN, LSTM, CNN-LSTM, Inception) to measure vital parameters. [Filed Patent Ref No. 201731038573]. The 99% accuracy was achieved in Respiration and Heart Rate measurement with 30-sec epochs. Deep learning models were trained on the Google Cloud Platform (GCP) with Tensorflow-GPU.Successfully achieved kernel changes into Android OS to read data on Qualcomm’s Snapdragon processor using SPI, I2C, and UART. Interfaced Java and custom module coded in C-Language in Android 6.0.Designed my own email marketing software for the marketing of SleepDoc. Used AWS Boto3 Python, AWS SES, SNS, SQS, AWS Lambda, DynamoDB, and PyQt5. Written code for automatic email and phone number extractor crawler.

Student

• Jul 2012 - Feb 2017

Teacher Assistant

• Jan 2013 - Jan 2017

It was a great experience to teach Embedded Systems. The objective of the teaching is the implementation of the adaptive filter on embedded hardware and software.Implementation of the DSP algorithms was done in Assembly language. Students get comprehensive knowledge of Embedded System and IoT.

M.Tech

• Jul 2012 - May 2014

Intern

• May 2011 - Jul 2011

Micro-Copter control system design Micro-Copter control system design