Ritika Mangla

Ritika Mangla

Masters in Computer Science

The University of Texas at Austin

About Me

I am a Master’s in Computer Science student at the University of Texas at Austin. I am currently advised by Dr. Jessy Li (co-advisor Dr. Greg Durrett). My research is focused on understanding discourse relationships using the linguistic framework of QuD (Question under Discussion). I am working on developing an evaluation framework for QuD parsers and exploring QuD ranking.

In addition to this, I have a keen interest in music. I am a music aficionado at heart and the days I seem to hit the wrong note, my guitar helps me harmonize. My days aren’t complete without an element of music.

You can find my Résumé here.

Interests
  • Natural Language Processing
  • Computational Linguistics
Education
  • Master's in Computer Science, May 2024 (expected)

    The University of Texas at Austin

  • B.Tech in Information Technology, 2022

    Veermata Jijabai Technological Institute

Experience

 
 
 
 
 
The University of Texas at Austin
Graduate Research Assistant
The University of Texas at Austin
Jan 2023 – Present Austin, TX
Developing the first linguistically driven framework for automatic evaluation of QuD (Questions under Discussion) parsers and also exploring QuD ranking. The applications include text simplification and summarization, guided text writing and improving conversational AI
 
 
 
 
 
Hewlett Packard Enterprise
Software Engineer Intern
Hewlett Packard Enterprise
May 2023 – Aug 2023 San Jose, CA
  • Automated the workflow of the namespace index module using C++ based customized gRPC services which targeted large workloads, thereby reducing manual testing hours by 60%
  • Integrated a standalone centralized Elastic-search based repository for storing IO workloads with the common feature library for improved search ability while reproducing specific customer use-cases or triaging performance issues
  • Contributed to 4 different code repositories by cross-functionally collaborating with 5 different teams and stakeholders
  • Received 2 HPE Star Points as a recognition of my work exceeding expectations
 
 
 
 
 
Goldman Sachs
Summer Analyst
Goldman Sachs
Jun 2021 – Jul 2021 Bangalore, India
  • Engineered an event creation and management platform, saving 10,000+ man hours for budgeting and approvals
  • Integrated new front-end and middleware with existing legacy backend which was used by 100+ internal departments
  • Designed test cases for unit testing and exposed to building the CI/CD pipeline for deploying production-level code
  • Collaborated cross-functionally with UI/UX teams and stakeholders to design a dashboard surfacing business insights
 
 
 
 
 
COE-CNDS Lab
Undergraduate Research Assistant
COE-CNDS Lab
Dec 2019 – Dec 2021 Mumbai, India
  • Developed a novel approach of ‘bot-score’ to detect bots in Twitter with an F1 of 97.3%
  • Performed feature engineering and statistical data analysis to understand the influence of bots in trending a hashtag
  • Examined the influence of bots by detecting communities with bot representations using the Louvain algorithm
  • Developed a FLASK based web-app to make an easy-go-access interface to ascertain a Twitter user’s bot probability
 
 
 
 
 
Yozu
Artificial Intelligence Intern
Yozu
Dec 2020 – Jan 2021 IIT Bombay, Mumbai, India
  • Developed an AI-based conversational bot using state machine approach with a reduced latency of 13%.
  • Devised an LSTM-based Question Matching algorithm for answering mid-conversation user queries.
 
 
 
 
 
TechnoPurple Tracking
Machine Learning Intern
TechnoPurple Tracking
May 2020 – Jul 2020 Mumbai, India
  • Implemented Laplacian filter to separate dark and blur images to examine site cleanliness for Emrill Services LLC(Dubai).
  • Engineered a template matching using OpenCV & OCR-based algorithm to determine success of a rally of Bhartiya Janta Party (the current ruling party of India), thereby resulting in reduction of false positive and false negative counts by 67%.

Projects

*
Bearing Fault Analysis

Bearing Fault Analysis

Estimated the remaining useful life of a bearing using Machine Learning. Achieved better prognostic performance with 98.2% accuracy as compared to SOTA techniques such as Kalman filter.

Cleangenix

Cleangenix

Devised a solution to improve city sanitation under Solid Waste Management Department using Neo4J and PostGreSQL. Developed an algorithm to identify trash from geotagged images using a Mask R-CNN model.

Startup Quest

Startup Quest

Created a social media bot which indexes 2000+ potential startup ideas in a database and a CRM dashboard to display it. Developed a news feed scraper to trace activities of the projects by web scraping data using Beautiful Soup.

Unsupervised Speech Summarization

Unsupervised Speech Summarization

Performed text summarization by clustering sentence embeddings trained to embed paraphrases together using K-Means. Improved the accuracy by using a Skip-Thoughts (GRU-RNN) model to preserve sequence of words in a sentence.

GeoTracking of Waste

GeoTracking of Waste

Devised a platform for BMC, the governing civic body, to view the real time location of garbage trucks and the amount of waste collected.Created a dashboard to view daily statistics and reports in order to formulate measures to optimize waste management

Languages and Frameworks

c
C++
html
HTML
python
Python
react
ReactJs
css
CSS
javascript
JavaScript
spring
Spring
flask
Flask
nodejs
NodeJS

Machine Learning Libraries

pandas
Pandas
numpy2
Numpy
matplotlib
Matplotlib
opencv
OpenCV
scikit_learn
Scikit Learn
keras
Keras
tensorflow
TensorFlow

Contact

Please feel free to reach to me for any query