Saurabh Nair
Applied Scientist at Wayve- Claim this Profile
Click to upgrade to our gold package
for the full feature experience.
-
English Professional working proficiency
-
Hindi Native or bilingual proficiency
-
Malayalam Native or bilingual proficiency
-
Marathi Native or bilingual proficiency
Topline Score
Bio
Credentials
-
Probabilistic Graphical Models - Representation
CourseraMar, 2020- Nov, 2024 -
Machine Learning
CourseraApr, 2015- Nov, 2024 -
Algorithms: Design and Analysis, Part 1
CourseraMar, 2015- Nov, 2024 -
Survey of Music Technology
CourseraNov, 2014- Nov, 2024
Experience
-
Wayve
-
United Kingdom
-
Software Development
-
100 - 200 Employee
-
Applied Scientist
-
Sep 2022 - Present
Vision-Action-Language modeling for Embodied AI Vision-Action-Language modeling for Embodied AI
-
-
-
Giant.AI
-
United States
-
Software Development
-
1 - 100 Employee
-
Deep Learning Research Engineer
-
Nov 2021 - Jul 2022
Neural Radiance Fields(NeRFs): 1. Explored various static, efficient NeRFs like Instant-Neural-Graphics-Primitives, Kilo-NeRFs, and Plenoxels 2. Sparse depth supervision from structure from motion for training efficiency 3. Worked on Pixtrack, a NeRF and feature metric alignment based object pose tracking pipeline Robotics: 1. Worked on generating adversarial scenarios for the reinforcement learning (A2C) or Imitation Learning based robot-manipulators 2. Explored+Demoed Orb-SLAM3 for head-pose tracking and pixel-stick Show less
-
-
-
Vicarious
-
United States
-
Industrial Automation
-
1 - 100 Employee
-
Robotics Engineer
-
Feb 2018 - Nov 2021
Object Placement (2019-2021): 1. Worked on robot behaviors for tight and loose fit insertion problems using force-torque sensors, perception, and motion planning 2. Built the software foundations for generic insertion pipelines for multiple products 3. Explored Visual Servoing(IBVS) techniques to bridge the gap of pose-based insertion and camera calibration errors 4. Worked on a patent pending idea for efficient pose based placement of objects 5. Designed, built and deployed a shelf-insertion pipeline for apparel in collaboration with Grey Orange robotics Sensing(2019) 1. Built an RGB-D camera using online sensor fusion, and bespoke calibration rigs 2. Built an automated camera-calibration and verification pipeline for intrinsics, extrinsics of RGBD cameras attached to robots. Planning and perception(2018-2019): 1. Updated the motion-planning stack to represent the planning scene with Google-Protobufs, built tools to easily configure the robot-cell constraints using RViz 2. Built an optimized version of RGBD-self filter to remove points from point-clouds that belong to the robot 3. Worked on outlier detection methods to clean pointclouds before being used to construct octomaps Show less
-
-
-
Carnegie Mellon University
-
United States
-
Higher Education
-
700 & Above Employee
-
Research Assistant
-
Sep 2017 - Dec 2017
Multiple Object Tracking for autonomous driving. 1. Vehicle Pose detection using L-shape fitting 2. Multiple object tracking using Hungarian association and Multiple hypothesis tracking 3. Outlier filtering using T, and J-linkage algorithms Multiple Object Tracking for autonomous driving. 1. Vehicle Pose detection using L-shape fitting 2. Multiple object tracking using Hungarian association and Multiple hypothesis tracking 3. Outlier filtering using T, and J-linkage algorithms
-
-
-
Magic Leap
-
United States
-
Computers and Electronics Manufacturing
-
700 & Above Employee
-
Robotics Intern
-
May 2017 - Jul 2017
Vision guided Camera Calibration: - Started by designing the use-case, requirements, and project timeline - Built an RRT* based safe visual servoing routine to calibrate an RGB camera with Charuco boards - Initially prototyped using Gazebo, ROS, and MoveIT! - Demoed the end2end solution on a Kuka-IIWA manipulator and UEye cameras by the end of the internship Vision guided Camera Calibration: - Started by designing the use-case, requirements, and project timeline - Built an RRT* based safe visual servoing routine to calibrate an RGB camera with Charuco boards - Initially prototyped using Gazebo, ROS, and MoveIT! - Demoed the end2end solution on a Kuka-IIWA manipulator and UEye cameras by the end of the internship
-
-
-
Cisco
-
United States
-
Software Development
-
700 & Above Employee
-
Software engineer
-
Aug 2014 - Jul 2016
- Software Development of the ASR1000 enterprise routers. - Development of PHY, Port-channel, platform infrastructure - Automated testing using TCL and Python - Productivity, and automation tools for the team - Software Development of the ASR1000 enterprise routers. - Development of PHY, Port-channel, platform infrastructure - Automated testing using TCL and Python - Productivity, and automation tools for the team
-
-
-
Center of Excellence (CoE), VJTI
-
Mumbai Area, India
-
Researcher
-
Jun 2013 - Aug 2013
- Mathematical modeling, path planning, and Computer Vision based decision-making for Robo-soccer between two humanoids - Published results at ACODS-16. - The working prototype was covered and broadcasted by Indian National TV Doordarshan Sahyadry - Mathematical modeling, path planning, and Computer Vision based decision-making for Robo-soccer between two humanoids - Published results at ACODS-16. - The working prototype was covered and broadcasted by Indian National TV Doordarshan Sahyadry
-
-
Education
-
2016 - 2017
Carnegie Mellon University
Master of Science - MS, Robotic-systems Development -
2010 - 2014
Veermata Jijabai Technological Institute (VJTI)
Bachelor of Technology (B.Tech.), Electronics and Communications Engineering
