Devi Sandeep Endluri

3645 Wellborn Rd, Unit #1724A, Bryan, TX 77801 · +1 (979) 739-3429 · dsandeep97@tamu.edu

I am currently a Masters student in the Department of Computer Science and Engineering at Texas A&M University.

I have worked in DATA Lab under the supervision of Prof. Xia (Ben) Hu. I have applied AutoML techniques to Natural Language Processing tasks such as Named Entity Recognition. My research interests include Machine Learning, Deep Learning and Natural Language Processing.

Prior to my Masters, I have work experience of 5 years as a Software Engineer in Qualcomm India Private Limited, Hyderabad, India.


Experience

Data Science Research Intern

Penn State University, State College, Pennsylvania
May 2020 - Aug 2020
  • Developed a fully automated end-to-end framework (ChartReader) to extract data from plots in scientific research papers
  • Classified charts from research papers using VGG Neural Networks with an accuracy of 84.01 across 13 chart categories
  • Used Amazon Rekognition text detection to detect text from chart images. Applied Computer Vision techniques using OpenCV to detect x-axis and y-axis ticks with an accuracy of 98% and 68% respectively, 83% accuracy in detecting legends and 42% in extracting data values in the testing corpus

Graduate Student Researcher (under Prof. Xia Ben Hu)

Data Analytics at Texas A & M (DATA) Lab, Texas A & M University
Jan 2020 - Dec 2020
  • Working on a pipeline utilizing AutoML to automatically search for a best neural model for Named Entity Recognition NLP task
  • Constructed Knowledge graphs based on 0.9M relations extracted from COVID-19 Open Research Dataset (CORD-19)

Senior Software Engineer

Qualcomm India Private Limited
July 2014 - July 2019
  • Facilitated design, development and maintenance of proprietary software to manage data connectivity of mobile devices in a smart and efficient way in terms of user experience.
  • Engaged in development and debug of AOSP (Android Open Source Project) in Android Connectivity domain
  • Spearheaded various IMS critical value-add features (G2L Tuneaway, Dual VoLTE) for Qualcomm chipsets
  • Interacted with multiple internal and external teams to develop features end-to-end. Experience with partnership and collaboration with customers, ecosystem providers and support, during all stages of software product life cycle
  • Strong involvement in Qualcomm MSM/MDM Chipset bring-up, debug, integration and commercialization
  • Awarded 5+ Qualstars, Orion Insta award in appreciation of outstanding contributions to Android Connectivity domain

Software Engineering Intern

Qualcomm India Private Limited
May 2013 - July 2013
  • Developed a command-line automation tool with Perl GUI Toolkit to validate presence of all critical non-volatile items in Android Phone memory.
  • Designed a consolidated platform to manage customer requests and software releases

Summer Research Intern (under Prof. Indranil Sengupta)

Indian Institute of Technology, Kharagpur
June 2012 - July 2012
  • Implemented a synthesis approach for reversible logic circuit based on theory of permutation cycles, using multile-control Toffoli gates with the main objective of educing quantum cost.

Education

Indian Institute of Technology Kharagpur

B.Tech in Computer Science
August 2010 - May 2014

GPA: 8.27 / 10

Texas A&M University

Masters in Computer Science
August 2019 - Present

GPA: 3.857 / 4

Skills

Programming Languages
Frameworks & Tools
Machine Learning
  • Classification
  • Regression
  • Feature Engineering
  • Data scraping
  • Data visualization
  • Data modeling

Projects

Animations in Cell Biology Learning Content
Developed interactive games and animations to improve biology learning experience of middle school students, and deployed them to the StepStone server.
Real-time COVID-19 Twitter Data Analytics
Real-time data analytics on COVID-19 over a Twitter Stream using Big Data Technologies of Hadoop Ecosystem such as Flume, Kafka and Spark Streaming. Flask web framework is used to display dashboards.
Deep Learning Image colorization based on U-Net
Implementation of Colorization using U-Net architecture and further experiments by modifying the model with different U-Net architectures with different number of down-sampling and up-sampling operations.
Abstractive Summarization using pre-trained encoders
Modified existing text summarization model with pre-trained BERTSUM encoder model and decoder architecture by introducing recurrence in model to improve better copying of source document.
Predictive models for flight arrival delay
Data models to predict flight delays for 3rd and 4th Quarters of 2019. Code for visualization dashboard using leaflet in R to visualize 2018 flight delay data.
Tacos and Burritos Data Visualization

Goldman Sachs Challenge, TAMU Datathon 2019

Built an interactive data visualization tool using Tableau to represent the Taco and Burrito data.
Linearization of Parallel Algorithms

B.Tech Project under Prof. Dipankar Sarkar, Indian Institute of Technology, Kharagpur

Implemented an algorithm to generate sequential code for parallel and multi-threaded programs with statements, predicates and multiple GOTO statements as programming constructs.

Publications


Achievements



Certifications

Get In Touch

Send me a message
+1 (979) 739-3429
dsandeep97@tamu.edu