APURV ASTHANA

https://www.linkedin.com/in/apurv-asthana-2a4a56a5

https://github.com/ApurvAsthana

*** **** **** ******, *** Angeles CA - 90007

213-***-**** ac4rmo@r.postjobfree.com

EDUCATION

University of Southern California, Los Angeles, CA May 2019 Master of Science Computer Science (Data Science) GPA: 3.7/4.0 National Institute of Technology, Raipur, India May 2014 Bachelor of Technology (Honors), Computer Science and Engineering TECHNICAL SKILLS

Programming Languages: C, C++, Java, Android, Python, SQL, JavaScript Platforms/Technologies: Windows, Linux, J2EE, Android, JIRA, Python Flask, Perforce, MySQL, NoSQL, MongoDB, Big Data, Map Reduce, IBM Watson, JSON, XML, HTML/CSS, NLTK. Coursework: Applied Natural Language Processing, Foundations of Artificial Intelligence, Database Systems, Information Retrieval and Web search engines. EXPERIENCE

Senior Software Engineer, Samsung R&D Institute Bangalore March 2016-June 2017

‘Snowball’ Project: Worked in a team for conceptualization and development of a voice controlled robot.

Modified Android platform to support requirements that couldn’t be implemented on it and removed unwanted modules.

Designed and implemented an application model using Java that system apps could use.

Developed a component using IBM Watson services (RESTful) for understanding user’s words and act accordingly. JifiCal: Developed a novel concept for creating calendar events by parsing user chat data or taking typed/spoken input, communicating with a server (using REST APIs) to analyze input and show relevant information to the user.

Implemented and deployed the app to the Android play store which was highly rated (5000+ downloads).

This service reduces the time (by more than 70%) and interactions (2 only needed) for creating a calendar event.

Worked in a team to engineer various Proof of Concepts(POCs) using Java and Android and carried out case studies to determine in which form the service could be provided and created the final product accordingly. Software Engineer, Samsung R&D Institute Bangalore June 2014-March 2016 English Input suggestions component in Calendar (S-Planner): Conceptualized an idea for improving User Experience in S-Planner which reduced event creation time by predicting and suggesting relevant inputs to the user and led a team to develop a POC on Android and Java and integrated it in S-Planner application. SmartManager application development: Developed SmartManager application which is used to optimize functioning of Samsung smartphones and is a built-in application in Samsung Android smartphones. PROJECTS

Chinese and English POS Tagger: Implemented a Hidden Markov Model based Part of speech tagger (on Python) for English and Chinese which learns model from a given tagged dataset and then uses the model to tag unseen data.

Review Classifier: Developed an agent on Python which can classify a given review as a positive or negative one using sentiment analysis. Trained using data set from UCI ML repository (Paper: 'From Group to Individual Labels using Deep Features', Kotzias et. al). The data set contains reviews from multiple websites.

Inverted Index: Created an inverted index list of words occurring in a set of English books from a subset of the Project Gutenberg corpus. I used Google cloud platform for storing the data and ran Hadoop Map reduce (written on Java) task over it to carry out the process and create an inverted index file.

Logic Inference Engine: Designed and implemented a Prolog-like inference engine on Java, which takes knowledge base clauses as input and then based on the knowledge base determines the truthfulness of user queries.

Adversarial Game Playing agent: Developed an intelligent agent in Python to select best moves within a time limit in two-player Reversi like game. Implemented it using minimax algorithm with alpha-beta pruning, as an academic project.

Data Mining Utility Tool: Programmed a utility tool on Java as an academic project which enables user to carry out tasks such as clustering, association rule mining and summarization on input data.