Social Media Data

SHIVAM SRIVASTAVA

***** ********** **** **, *****, Florida 33612

813-***-**** *******@***.*** LinkedIn

EDUCATION

University of South Florida Tampa, FL Aug. 2019 - May 2021 Master of Science in Computer Science

GPA: 3.95

JSS Academy of Technical Education Noida, India Aug. 2012 - June 2016 Bachelor of Technology in Computer Science & Engineering GPA: 3.75

PUBLICATION

Recognizing Emotion in the Wild using Multimodal Data S. Srivastava, S. Aathreya, S. Hinduja, Sk R. Jannat, H. Elhamdadi, and S. Canavan. International Conference on Multimodal Interaction, 2020 [LINK] PROGRAMMING SKILLS

Language Python, Java, Javascript

Databases MySQL, PostgreSQL, MongoDB, Redshift(AWS) AWS Services Lambda, Kinesis, SQS, SNS, S3, EC2, Athena, EMR, Glue Web Development Django, React JS, HTML/CSS

Big Data Hadoop, Spark, Kafka, Air ow

Deep Learning TensorFlow, PyTorch, Numpy

Deployment Nginx, Gunicorn, Ansible

WORK EXPERIENCE

University of South Florida Jan. 2020 - Present

Teaching Assistant and Student Researcher

Teaching Assistant for Introduction to Theory of Algorithms (COT 6405) instructed by Dr. Shaun Canavan.

Ongoing Research: Automatic emotion and facial expression detection using non verbal cues. Delhivery June 2017 { June 2019

Software Engineer

Responsible for building a scalable enterprise level data warehouse enabling teams to get historical as well real-time state of all the packages and courier vehicles across 17,500+ centers in India.

Managed a multi-node Redshift cluster, handling 2000+ queries serving di erent dashboards daily. Separat- ing OLAP and OLTP queries reduced production downtime by 2X and also improved the average time for analytical queries by 30%. Collaborated with the Infrastructure team frequently for ensuring scalability, security and availability of the cluster and other resources.

Created an ETL pipeline using AWS Big Data Ecosystem and Apache Air ow that validates, clean and store the data. Implemented an Apache Spark job that converts data on S3 in row storage JSON format to columnar storage Parquet format which reduced the storage cost by 80% on S3 and approximately 70% data scans on queries, reducing Athena’s (AWS’s query service) per-TB-scanned billing cost.

Implemented unit testing using Pytest and an o ine integration testing pipeline using localstack/moto library which removed the dependency of creating and maintaining resources for test environment, thus cutting total resource’s cost by 25%. Concurrently, assisted in designing architecture and schema for distributing, sorting and storing data such that it tunes query performance. Neuron June 2016 { May 2017

Software Engineer

Created and managed Neuron’s developer platform which o ered multiple natural language processing based services to users according to the subscription opted. Responsible for designing and developing server side RESTful services using Django REST Framework.

Created a scalable and interactive Chatbot and it’s admin panel in 2 months to build and store surveys that can be shared easily on multiple social media platforms and sites, replacing the usual form type questionnaire format for clients. Used Django, React JS as the web frameworks and MySQL, MongoDB as the databases.

Followed agile methodology with automated unit testing and CI/CD using Jenkins to ensure a seamless deployment of code on production and Git/BitBucket as the version control tool.

Worked closely with clients, AI team and Senior Developers for requirement gathering, planning and imple- mentation of the projects with focus on high availability, low latency and scalability. INTERNSHIPS

Neuron Dec. 2015 - Apr. 2016

Integrated Hindi news Ad recommendation system with clients applications to fetch and display other related news as well along with the main news.

Integrated a JavaScript library, ngerprint.js, on client’s machine for anonymous browser detection and hence throttling user based on unique ngerprint/ID.

Algoscale May 2015 - July 2015

Devised a web based analytics tool to fetch and analyze the keywords related to a YouTube channel.

Developed an admin portal for clients using Django’s MVT architecture using which clients can get the result in their desired text format.

FratMart Sep. 2014 - Mar. 2015

Developed an online peer-to-peer marketplace for buying and selling used stu (books, study material, novels, electronics etc), nding ats/ atmates or arrange carpools to college.

Responsible for the development of the whole product from scratch, managing the server, database manage- ment and other both backend and frontend related tasks. SELECTED PROJECTS

JSS Infoconnect: Revamped the old PHP-based web architecture of Student-Faculty interaction platform using modern framework and libraries Django and Angular, serving 10,000+ Daily Active Users

Food Genie: Android based app for nding ratings of food items per restaurant. Used Natural Language Processing on user and critic reviews of restaurants to analyze the frequency and sentiment of users towards a particular item and calculate the rating.

PyURL Shortener: Create short and custom URLs that can be easily shared. SELECTED COURSEWORK

Digital Image Processing, Computer Vision, Pattern Recognition, Introduction to AI, Introduction to Robotics, Distributed Systems, Design and Analysis of Algorithms, Theory of Computation, Software Engineering ACHIEVEMENTS AND EXTRA CURRICULAR

3rd place, Eighth Emotion Recognition in the Wild Challenge (EmotiW 2020).

Winner, Web Developemt @ Zealicon Tech fest, JSS Noida.

Runner-up, GDG-Lingaya’s App Development Hackathon, Delhi.

Organiser, Google Developer Group, JSS Noida.

Organiser, Hackathon 2015, JSS Noida.

Web Head, Microsoft Mobile Innovation Lab, JSS Noida.

Contact this candidate