Sanjana Garg

Sanjana Garg

Masters student in Computer Science

Georgia Institute of Technology

Biography

I am a Masters’ student in Computer Science at Georgia Institute of Technology. I am working with Prof. Munmun De Choudhury, as part of SocWeb Lab where my research is focussed on analyzing drug abuse discourse on social media. My interests lie in natural language processing and social computing.

I also worked as a Software Developer at Rivigo for 2 years. I worked on several challenging projects here including Rivigo Maps, search engine and notification service.

Interests

  • Natural Language Processing
  • Social Computing
  • Machine Learning

Education

  • Masters in Computer Science, 2021

    Georgia Institute of Technology

  • BTech in Computer Science and Engineering, 2017

    Indian Institute of Technology Kanpur

  • Minor in Linguistics, 2017

    Indian Institute of Technology Kanpur

Experience

 
 
 
 
 

Machine Learning Intern

Bloomberg LP

May 2020 – Jul 2020 New York, USA
  • Built a pipeline of NLP based solutions to answer Excel related queries (around 8% of total queries) on Bloomberg Terminal’s HelpDesk.
  • Finetuned a BERT classification model for identifying relevant queries with an F1 score of 0.7.
  • Integrated a grammar-based parser that output an Excel formula, which can be directly suggested to the user as an answer.
 
 
 
 
 

Software Development Engineer I and II

Rivigo

Jun 2017 – Jul 2019 Gurgaon, India
  • Revamped rule-based ETA prediction with AI-based, using ML models, honoring constraints and crunching millions of GPS data points, building a powerful service called Rivigo Maps.
  • Worked with the Vyom team to develop a Pilot app for fleet owners for a seamless experience of tracking, refueling, and reconciliation of payments.
  • Decreased search latency by 10 times by developing a search engine leveraging ElasticSearch, powering search across all platforms.
  • Designed and developed a scalable notification service supporting features like tracking, queuing, filtering and heavy attachments.
 
 
 
 
 

Algorithmic Trading Intern

Kivi Capital

May 2016 – Jun 2016 Gurgaon, India
  • Designed and back-tested high-frequency pair trading strategies with time-adaptive approaches.
  • Experimented with evolutionary algorithm approaches for exploiting the volume imbalance.
 
 
 
 
 

Cloud Computing Intern

Reliance Jio

May 2015 – Jun 2015 Gurgaon, India
  • Implemented a module that gathered and rotated the service logs at a given interval.
  • Studied Ceilometer project of Openstack.

Publications

Agent based simulation of the evolution of society as an alternate maximzation problem

Understanding the evolution of human society, as a complex adaptive system, is a task that has been looked upon from various angles. In …
Amartya Sanyal, Sanjana Garg, Asim Unmesh
PDF Code Slides

Contact