Nitish Srivastava
Machine learning researcher Nitish Srivastava connects neural-network research with autonomous systems. He earned a computer science degree at IIT Kanpur in 2011, completed a PhD at the University of Toronto in 2017, and coauthored papers on dropout and video representation learning. He co-founded Clarevision Research and worked at Apple on embodied agents and autonomous perception before co-founding Vayu Robotics, whose 2024 delivery robot paired a mobility foundation model with passive sensing. Since August 2025, he is Senior Director of Autonomy at Serve Robotics; IIT Kanpur's alumni association named him among its 2026 Young Alumni Awardees.
- Now at
- Serve Robotics
- Works
- 36
- Citations
- 44,752
- h-index
- 22
Publication numbers from OpenAlex, as of 22 October 2026.
Timeline
-
2011–2017
IIT Kanpur and a Toronto PhD
He earns a BTech in computer science at IIT Kanpur in May 2011 and completes a computer science PhD at the University of Toronto in 2017. In Toronto's Machine Learning Group, he studies with Geoffrey Hinton and Ruslan Salakhutdinov.
-
2014
The dropout method
With Geoffrey Hinton, Alex Krizhevsky, Ilya Sutskever and Ruslan Salakhutdinov, he coauthors Dropout: A Simple Way to Prevent Neural Networks from Overfitting in 2014. The method randomly drops network units and their connections during training to reduce overfitting.
-
2015
Video representations with LSTMs
In 2015, with Elman Mansimov and Ruslan Salakhutdinov, he coauthors Unsupervised Learning of Video Representations using LSTMs. Its encoder-decoder model learns representations of video sequences and predicts future sequences.
-
2015–2016
Clarevision for autonomous driving
From 2015 to 2016, he co-founds Clarevision Research with Ruslan Salakhutdinov and Charlie Tang. The company develops perception neural networks for autonomous driving.
-
2017–2022
Apple research and robotic control
At Apple, he works as a Staff Research Scientist from 2017 to 2022, applying self-supervised and reinforcement learning to embodied agents and building perception models for autonomous technologies. In 2021, he coauthors Robust Robotic Control from Pixels Using Contrastive Recurrent State-Space Models; its recurrent model predicts the next observation and supports robot control under camera, background and color distractions.
-
February 2022–August 2025
Co-founding Vayu Robotics
From February 2022 to August 2025, he serves as CTO and co-founder of Vayu Robotics. In October 2023, the company describes a foundation model for mobility and a low-cost sensing technology that could replace lidar in some mid-range applications; its March 2024 announcement identifies him as Vayu's CTO.
-
July 2024
Vayu's AI delivery robot
In July 2024, Vayu debuts an on-road delivery robot that combines a transformer-based mobility foundation model with a passive sensor. The robot operates without lidar or pre-mapping the roads it will travel.
-
Since August 2025
Senior Director at Serve Robotics
Since August 2025, he is Senior Director of Autonomy at Serve Robotics, where he leads the machine-learning team that builds intelligence for safe navigation in complex urban environments. The IIT Kanpur Alumni Association includes him among its Young Alumni Awardees in 2026.
Career
- Since 2025Serve RoboticsSenior Director of Autonomy
- From 2023Vayu RoboticsCo-founder, CTO
- 2022–2024Google
- 2017–2022AppleStaff Research Scientist
- 2017–2020Cornell University
- 2015–2016Clarevision ResearchCo-founder
Education
- 2011Indian Institute of Technology KanpurBachelor's degree in computer science
- University of TorontoPhD
Most-cited works
- Dropout: a simple way to prevent neural networks from overfittingWith Alex Krizhevsky, Geoffrey Hinton and Ruslan Salakhutdinov
- Improving neural networks by preventing co-adaptation of feature detectorsWith Alex Krizhevsky, Geoffrey Hinton and Ruslan Salakhutdinov
- Unsupervised Learning of Video Representations using LSTMsWith Ruslan Salakhutdinov
- Multimodal Learning with Deep Boltzmann MachinesWith Ruslan Salakhutdinov
- Improving Neural Networks with Dropout
Sources and credits
Photo: We have no published photo of Nitish Srivastava yet. Send us a free-licensed photo.
- Nitish Srivastava
- CV - Nitish Srivastava
- Dropout: A Simple Way to Prevent Neural Networks from Overfitting
- Unsupervised Learning of Video Representations using LSTMs
- About me - Nitish Srivastava
- Robust Robotic Control from Pixels Using Contrastive Recurrent State-Space Models - Apple Machine Learning Research
- Vayu Robotics Emerges from Stealth with $12.7 Million in Seed Funding from Khosla Ventures and Lockheed Martin Ventures
- Vayu Robotics News
- Vayu Robotics debuts the world’s first on-road Delivery Robot powered by an AI Robotics foundation model
- Young Alumnus Awards




