- I am a final year student at BITS Pilani, Pilani Campus pursuing my Bachelors in Electrical and Electronics Engineering.
- I am currently an undergraduate Research Assistant at the Maryland Information and Network Dynamics (MIND) Lab at the University of Maryland, College Park, MD, USA.
- I was previously a Data Science Intern and an onsite Data Technician at TurnoutNow LLC, a cloud-based data analytics firm based in Lancaster, Pennsylvania in the United States.
- I am committed to doing research projects which have the potential of making a positive impact on humanity. My areas of interest span across Geopositioning, Cloud Computing, Operating Systems, Machine Learning, Information Retrieval and NLP. I have worked on various projects across these domains.
- I am interested in working on machine learning models to solve challenging problems across various domains. I yearn to learn something new and add to my knowledge of machine learning and data science.
- When I am not engrossed in academic activities, I can be found exploring recent machine learning techniques, catching up with latest technology and listening to music.
- Undergraduate research thesis under Prof. Ashok Agrawala, Founder and Director of Maryland Information and Network Dynamics (MIND) Lab , University of Maryland, College Park, Maryland, USA.
- My work here focuses on analyzing the spread of COVID-19 and Flu virus using breathing and location data collected from Spire tags.
- This is a part of the PROMETHEUS Project in collaboration with the School of Public Health.
- Designed algorithms to efficiently determine proximity pairs for contact tracing on a data set of more than 500K data points. Generated proximity pairs between two users within a distance of 3 meters and within a span of 15 minutes from each other.
- Developed algorithms for time series analysis for segmentation of breathing data to identify singular breathing events and anomalies like talking, sneezing, laughter, cough etc.
- I was a Research Intern under Amandeep Gill, Executive Director, UNSG's Panel on Digital Cooperation.
- The International Digital Health and AI Research Collaborative (I-DAIR) Project seeks to advance the UN Secretary General's (UNSG) High-level Panel on Digital Cooperation's recommendations related to digital health, and targets set at the World Health Organization (WHO) on universal and quality health coverage.
- Researched the role of micro-narratives as proxy variables to fill in missing data and used natural language techniques to study the social, health, and impacts of the COVID-19 crisis on various sections on the society.
- Analyzed how Big Data from 25000+ IoT BLE Beacons is created and captured by over 900+ IoT data capture devices (Session App).
- Used vert.x-core framework for capturing and storing this Big Data into Redis. Used Java to process it in real-time and MongoDB for archiving.
- Used R to generate real-time recommendations based on the attendee's live location on the show floor.
- The data was then analyzed in Power BI Dashboards which displayed an overall summary of the event with natural language insights.
- Understood the interaction of real-time natural language generation tools with live data connections and generated narratives for end users. These narratives provided controlled insights based on user inputs in concise natural language text.
Implemented a neural networkfor utilizing information in SEC 8-K forms for predicting the movement of the S&P 500 index.
Used BERT for capturing the contextual information in the form of two methods: Masked Language Modeling (MLM) and Next Sentence Predicting (NSP).
DigiYatra is an initiative by the Ministry of Civil Aviation, Government of India that aims to revolutionize air travel in the country.
Explored the implications of Machine Learning and Artificial Intelligence as an asset to this program.
Studied the security and privacy challenges involved in implementing technologies like facial recognition, passenger tracking, recommender systems, etc.View Report
Used information from DNS queries to predict a DDoS attack from Darknet data from the Center for Applied Internet Data Analysis (CAIDA) supercomputer servers of the University of California, San Diego (UCSD). for research purposes.
Implemented Python scripts for feature extraction (like TTL, IP length, Packet Count, etc.) and CAIDA's internal tool, corsaro for large scale analysis of trace data.
Used vector quantization algorithms including K-means and EM on the extracted features to predict DDoS attacks.View Source Code
Built and implemented a Cross-Lingual Document Translator using Statistical Machine Translation Model for Dutch and English documents.
Implemented the IBM Model 1 for lexical translation and alignment along with the Expectation Maximisation (EM) algorithm to build and train the translator for a corpus of over 2 million sentences.
Used ensemble techniques like Bagging and Boosting to improve the precision and recall of the CLIR.View Source Code
Designed and implemented using Assembly Level Language in Proteus to test 7408, 7486, 7432 IC chips for correct logical functioning as described in the truth and/or function table for any given IC chip.
This involved interfacing peripherals like RAM, ROM, PPI, Buffers, Decoders, Latches and LCD Display to the 8086 microprocessor.
Project was simulated successfully using Proteus.View Code