Subendhu Rongali

Applied Scientist

Amazon Alexa AI, New York


I am an Applied Scientist at Amazon Alexa AI. I work on Alexa’s language understanding engine and my research interests include semantic parsing and speech/conversational AI. I got my PhD in Computer Science from UMass Amherst in 2022. I worked with Prof. Andrew McCallum in the Information Extraction and Synthesis (IESL) Lab and my dissertation focused on improving low resource language understanding in voice assistants.

When I’m not working on research, I’m mostly playing on my PS5 or cheering for India in cricket. I also enjoy tennis and cooking with my wonderful and brilliant wife Emily First.


  • Machine Learning/Natural Language Processing
  • Semantic Parsing
  • Speech/Conversational AI


  • M.S/Ph.D in Computer Science, 2022

    University of Massachusetts Amherst

  • B.Tech in Computer Science and Engineering, 2014

    Indian Institute of Technology Madras


Jan 2023: Our paper, Low-Resource Compositional Semantic Parsing with Concept Pretraining, was accepted at EACL 2023.
Sep 2022: I started work as an Applied Scientist at Amazon Alexa AI.
Jul 2022: I successfully defended my doctoral dissertation at UMass Amherst!
May 2022: I successfully proposed my doctoral dissertation ( pdf, slides) at UMass Amherst. My committee consists of Andrew McCallum, Mohit Iyyer, Andrew Lan, and Konstantine Arkoudas.
Apr 2022: Our paper, Training Naturalized Semantic Parsers with Very Little Data, was accepted at IJCAI 2022.
Feb 2021: I presented a poster on my internship work at AAAI 2021.
Dec 2020: My internship work at Alexa in 2020, Exploring Transfer Learning for End-to-End Spoken Language Understanding, was accepted at AAAI 2021.
Sep 2020: Our paper, Unsupervised Parsing with S-DIORA: Single Tree Encoding for Deep Inside-Outside Recursive Autoencoders, was accepted at EMNLP 2020.
Sep 2020: Our paper, Compressing Transformer-Based Semantic Parsing Models using Compositional Code Embeddings, was accepted at EMNLP Findings 2020.
Feb 2020: Our paper, Learning Latent Space Representations to Predict Patient Outcomes, was accepted at JMIR.
Feb 2020: My internship work on improving Alexa commands was featured on Venturebeat.
Jan 2020: My internship work at Alexa in 2019, Don’t Parse, Generate! A Seq to Seq Architecture for Task-Oriented Semantic Parsing, was accepted at WWW 2020.
Nov 2019: I presented a poster about my summer internship work at Amazon Research Day 2019.
Jul 2019: I was awarded the Sudha and Rajesh Jha Scholarship for 2019 at CICS, UMass Amherst.

Work Experience


Applied Scientist

Amazon Alexa AI

Sep 2022 – Present New York City, NY

Applied Scientist Intern

Amazon Alexa AI

May 2019 – Sep 2021 New York City, NY
Interned during summers 2019, 2020, and 2021.

Software Developer

Epic Systems Corp.

Oct 2015 – Aug 2017 Madison, WI

Research Software Engineer

IBM Research

Sep 2014 – Sep 2015 Bangalore, India

Research Intern

Adobe Advanced Technical Labs

May 2013 – Aug 2013 Bangalore, India


Software Engineering, COMPSCI 320

Programming Methodology, COMPSCI 220

Artificial Intelligence, COMPSCI 383


Low Resource Language Understanding in Voice Assistants

This was my primary area of research with Prof. Andrew McCallum and the focus of my dissertation. We worked on improving low resource …

DIORA: An unsupervised tree parser and LM

In the past, I worked work with Prof. Andrew McCallum and Prof. Mohit Iyyer on improving DIORA (Deep Inside Outside Representations …

Machine Learning in Healthcare

This was my primary area of work with Prof. Hong Yu. We are working on a number of ML and NLP applications in healthcare. One line of …

Evidence Inference in Long Documents

This was my final class project for COMPSCI 692A: Automated Knowledge Base Construction. We tackle the problem of selecting evidence to …

Proof completion in the formal verification language - Coq

Formal verification is an important field in Software Engineering and in Computer Science in general. With increasingly complex …

High Confidence Off-Policy Improvement

This was our final course project for COMPSCI 687: Reinforcement Learning. This project simulates the application of RL to a real …

Gendered Pronoun Coreference Resolution

This was our final course project for COMPSCI 690D: Advanced Natural Language Processing. Coreference resolution is the task of …

Influence Functions in Machine Learning Tasks

This was our final course project for COMPSCI 689: Machine Learning. Machine learning models are generally complex and it is difficult …