Profile Picture

Arnav Thareja

Hi! I am a student at the University of Washington studying Computer Science and Mathematics.

A

r

n

a

v

T

h

a

r

e

j

a

Experience

Software Engineer Intern

Oracle Cloud Infrastructure (June 2022 - September 2022)

  • Working on the Virtual Machines Efficiency team within Oracle Cloud Infrastructure (OCI) Compute
  • Designed and built a system to monitor usage of reserved compute resources and identify resources to be reclaimed
  • Created internal usage metrics, dashboards, and alarms using Java and Oracle Monitoring Query Language (MQL)

Undergraduate Researcher

Personal Robotics Lab (May 2021 - Present)

  • Working on multi-agent autonomous navigation and task allocation with MuSHR cars
  • Designed and implemented algorithms for non-holonomic multi-agent navigation with optimal task allocation in C++
  • Extended existing control algorithms to the multi-agent domain to eliminate agent collisions and improve robustness
  • Built ROS (Robot Operating System) wrappers around algorithms to enable easy interfacing with existing systems
  • Sped up robot trajectory comparison framework by 50x by directly analyzing ROS bags through the rosbag Python API
  • Demonstrated and tested system capabilities and translation to real-world environments on physical robots

Software Engineer, Autonomous Navigation Subteam

Husky Robotics (October 2020 - October 2021)

  • Created robot pathfinding and autonomous navigation algorithms for a prototype Mars rover using C++
  • Integrated ROS2 into codebase using nodes and topics for navigation plan visualization
  • Defined and implemented a navigation algorithm to locate targets based on approximate GPS coordinates
  • Designed patterns for driving between two posts given GPS coordinates of the center
  • Leveraged Docker for CI (continuous integration)

Projects

Chess

A chess game that can be played in the terminal with minimax agents to play against. Minimax agents use alpha-beta pruning alongside a heuristic-based iterative deepening depth first search algorithm and memoization to improve runtime. Written in Java.

See the source code here.

Angles

A Chrome extension that suggests news articles of opposite bias when a news website is visited. Made with JavaScript, jQuery, and Google Cloud Natural Language.

Submission to Dubhacks 2020, placed in the top 3 in the Newsprint track.

See the Devpost submission here and see the source code here.

Yearbook 2020

A web app designed to keep students connected during COVID-19 quarantines. Each student has a virtual yearbook page that others can sign, and they are able to sign others' yearbook pages. Made with JavaScript, HTML, CSS, and Google Firebase.

Try it at https://yearbook-hhs.web.app and see the source code here.

More on GitHub!

See more of my projects on my GitHub!

https://github.com/arnavthareja

Coursework

Selected Coursework

  • Machine Learning (CSE 546)
  • Reinforcement Learning and Optimal Control (CSE 579)
  • Deep Learning (CSE 599)
  • Algorithms (CSE 421)
  • Autonomous Robotics (CSE 478)
  • Computer Vision (CSE 455)
  • Databases (CSE 344)

Planned Coursework

  • Deep Robotic Learning (CSE 599)
  • Modern Algorithms (CSE 422)
  • Quantum Computation (CSE 490)
  • AI (CSE 473)
  • Linear, Nonlinear, and Discrete Optimization (MATH 407-409)

Resume

Open my resume in another tab

Download my resume