Curriculum Vitae

Education

Ph.D in EECS, MIT, 2029 (expected)
S.M. in EECS, MIT, 2026 (expected)
B.S. in EECS, UC Berkeley, 2024

Work experience

2024 to Present: Researcher at Improbable AI Lab (CSAIL)
- Researching continual learning, foundation models, unsupervised training, and robot learning.
- Supervisor: Pulkit Agrawal
2020 to 2024: Researcher at Robotic AI and Learning Lab (BAIR)
- Researched deep reinforcement learning, unsupervised pre-training, and robot autonomy. Assisted with robotic learning research, carrying out experiments, writing code, and contributing to writing papers. Proposed soft-prompt tuning task adaptation for general robot navigation. Created a hybrid control system with deep imitative models and sensor fusion to improve out-of-distribution testing. Implemented a risk-aware DAgger-style training system with an MPPI controller for autonomous vehicle racing.
- Supervisor: Sergey Levine
Summer 2023: Research Intern at Center for Human-Compatible AI
- Studied human-inspired exploration methods and skill discovery training for RL. Proposed an approach for using Laplacian representations and skill-centric curiosity for acquiring a low-level policy. Created a value function space skill-chaining method for affordance-aware exploration and hierarchical planning.
- Supervisor: Stuart Russell
2020 to 2023: Researcher at Redwood Center for Theoretical Neuroscience
- Duties includes: Researched interpretable unsupervised capsule representations for part-object hierarchies. Developed variations of Lie Sparse Coding and Flow Capsules and created pooled deformation groups with RBMs. Led a team of 5 and created 4 weeks of course content, lectures, and assignments for an AI Perception MOOC on EdX.
- Supervisor: Bruno Olshausen
Summer 2022: Software Development Engineer Intern at Amazon
- Duties includes: Worked on Amazon Web Services Elastic Block Store to optimize big data Spark and Hadoop jobs for S3 snapshot analysis. Developed a Spark App to shard large parquet datasets for distributed compute load-balancing across multiple EC2 nodes. Achieved 300% performance gain with my multi-cluster system to provision and launch Elastic MapReduce cluster jobs
- Supervisor: Richa Mishra
Summer 2021: Summer Research Fellow at NASA, Caltech
- Duties includes: Developed a robotic visual perception framework for high speed planning which increased time-until-collision by 110%. Wrote a third-authored internal paper with digital signal processing results on Random Forest to classify voice activity. Presented real and simulated results for our goal-directed methods which improved rate of navigation success by 45%.
- Supervisor: Ali Agha-Mohammadi

Publications

Correct Me If I’m Wrong: Language Guided Exploration for VLAs

Sunshine Jiang, John Marangola, David Zhang, Nitish Dashora, Pulkit Agrawal, Zhang-Wei Hong. (2026). "Correct Me If I’m Wrong: Language Guided Exploration for VLAs." Reinforcement Learning for Vision-Language-Action Models Workshop.

Learning More from Less: Reinforcement Learning from Hindsight

Iris Xu, Sunshine Jiang, John Marangola, Nitish Dashora, Richard Li, Thomas Liu, Zexue He, Yuheng Zhi, Alex Pentland, Pulkit Agrawal, Zhang-Wei Hong. (2026). "Learning More from Less: Reinforcement Learning from Hindsight." Reinforcement Learning for Vision-Language-Action Models Workshop.

Prompt-Driven Exploration

Sunshine Jiang, John Marangola, David Zhang, Raghuram Kowdeed, Ruiyang Luo, Nitish Dashora, Richard Li, Pulkit Agrawal, Zhang-Wei Hong. (2026). "Prompt-Driven Exploration." arXiv preprint arXiv:2607.08837.

DexWrist: A Robotic Wrist for Constrained and Dynamic Manipulation

(2025). "Martin Peticco, Gabriella Ulloa, John Marangola, Nitish Dashora, Pulkit Agrawal." Under Review International Conference on Robotics and Automation 2026.

ViVa: Video-trained Value Functions for Guiding Online RL from Diverse Data

Nitish Dashora, Dibya Ghosh, Sergey Levine. (2025). "ViVa: Video-trained Value Functions for Guiding Online RL from Diverse Data." arxiv.

ViNT: A Foundation Model for Visual Navigation

Dhruv Shah*, Ajay Sridhar*, Nitish Dashora*, Kyle Stachowicz, Kevin Black, Noriaki Hirose, Sergey Levine. (2023). "ViNT: A Large-Scale, Multi-Task Visual Navigation Backbone with Cross-Robot Generalization." In the proceedings of 2023 Conference on Robot Learning (CoRL).

Imitative Models for Passenger-Scale Autonomous Off-Road Driving

Nitish Dashora, Sunggoo Jung, Dhruv Shah, Valentin Ibars, Osher Lerner, Chanyoung Jung, Rohan Thakker, Nicholas Rhinehart, Ali-akbar Agha-mohammadi. (2022). "Imitative Models for Passenger-Scale Autonomous Off-Road Driving." Presented at 2023 International Conference on Intelligent Robots and Systems (IROS).

Hybrid Imitative Planning with Geometric and Predictive Costs in Off-road Environments

Nitish Dashora*, Daniel Shin*, Dhruv Shah, Henry Leopold, David Fan, Ali Agha-Mohammadi, Nicholas Rhinehart, Sergey Levine. (2022). "Hybrid Imitative Planning with Geometric and Predictive Costs in Off-road Environments." In the proceedings of 2022 International Conference on Robotics and Automation (ICRA).

Talks

Imitation Learning for Passenger-scale Control

October 03, 2023

Conference Proceedings Talk at International Conference on Intelligent Robots and Systems 2023, Detroit, MI

Hybrid Imitative Planning

May 24, 2022

Conference Proceedings Talk at International Conference on Robotics and Automation 2022, Philadelphia, PA

Service and Leadership

Officer of the Machine Learning Club at UC Berkeley (2020-23)
Officer of the Engineering Student Council at UC Berkeley (2020-21)

Nitish Dashora