CV

Ph.D. in Computer Science

University of Texas at Austin (UT Austin)

• Advisor: Kristen Grauman
• Research Interests: Robotics, Computer Vision, Representation Learning, Reinforcement Learning

M.Eng. in Electrical Engineering and Computer Science

Massachusetts Institute of Technology (MIT)

• Concentration: Artificial Intelligence
• Thesis: "Landslide Susceptibility Prediction Adaptive to Triggering Events"
  • Research Group: MIT Environmental Solutions Initiative
  • Advisor: John E. Fernández

S.B. in Computer Science and Engineering

Massachusetts Institute of Technology (MIT)

• GPA: 4.8/5.0
• Groups: MIT InterVarsity, Black Student Union, African Students Association

AMAGO-2: Breaking the Multi-Task Barrier in Meta-Reinforcement Learning with Transformers

Jake Grigsby, Justin Sasek*, Samyak Parajuli*, Daniel Adebi*, Amy Zhang, Yuke Zhu Conference on Neural Information Processing Systems (NeurIPS), 2024

• * = equal contribution

UT Austin Computer Vision Group

Graduate Researcher

• Working under Prof. Kristen Grauman developing better methods for video understanding tasks, such as procedural understanding and video alignment.

Machine Intelligence through Decision Making and Interaction Lab

Graduate Researcher

• Worked under Prof. Amy Zhang in the MIDI Lab developing better representations for training robots and other reinforcement learning agents to perform downstream tasks.

MIT Environmental Solutions Initiative (M.Eng. Thesis)

Graduate Research Assistant

• Designed and implemented various machine learning models to analyze and predict landslide susceptibility from LiDAR and satellite data.

MIT Computer Science and Artificial Intelligence Laboratory

Graduate Research Assistant at Kellis Lab

• Experimented with using graph variational autoencoders to learn representations for personalized Bayesian Networks in computational biology settings.

MIT Computer Science and Artificial Intelligence Laboratory

Undergraduate Researcher at Distributed Robotics Laboratory

• Developed and trained transformer-based deep reinforcement learning models to teach a fixed-wing drone how to fly in a virtual environment while accomplishing subgoals.

MIT Media Lab

Undergraduate Research Assistant in Camera Culture Group

• Studied the effectiveness of graph neural network-based simulations to build more effective forecasting systems.

IBM Research

AI Research Scientist Intern

• Used reinforcement learning to fine-tune large language models to play the word game Taboo. Worked on the Trustworthy AI team under Kush Varshney.

Google

Software Engineering Intern

• Experimented with using state-of-the-art NLP models for call transcript summarization.

Google

Software Engineering Intern

• Created an offline pipeline for the Google Lens team to extract labels from random forest decision tree models.

Google

Student Training in Engineering Program (STEP) Intern

• Contributed to a web application designed to help people learn various, user-chosen topics efficiently.

IBM

Software Engineering Intern

• Developed a program that created and maintained product representations that sellers use to sell to IBM Clients.

CS 373 - Software Engineering (UT Austin)

Graduate Teaching Assistant

CS 371P - Object-Oriented Programming (UT Austin)

Graduate Teaching Assistant

6.046 - Design and Analysis of Algorithms (MIT)

Grader and Tutor

6.036 - Introduction to Machine Learning (MIT)

Lab Assistant

• National GEM Consortium Fellowship