An assistive mobile manipulator, architected and deployed from scratch
At Sielo I own the software architecture for an assistive mobile-manipulation platform, from the first URDF through deployment on physical hardware. I designed the full ROS 2 system: motion planning, whole-robot control, RGB-D perception, mobility, teleoperation, and operator tooling, with interface boundaries clean enough that the simulation, planning, and hardware layers evolve independently.
01
Digital twin first
Every behaviour passes through a simulation-first validation pipeline: the digital twin verifies kinematics, controller scheduling, trajectory tracking, and robot-state synchronization before anything touches hardware. The pipeline is simulator-agnostic (Gazebo, Isaac Sim, MuJoCo), and closing the sim-to-real gap is treated as a core engineering loop, refining models and controllers against telemetry from the physical robot until simulated and real behaviour converge.
02
Down to the bus
I built the hardware abstraction layer from scratch: custom C++ ros2_control interfaces and actuator drivers bridging MoveIt 2 trajectory execution to CAN-based arm and mobility actuators, with explicit attention to control-loop rates, bus timing, and fault behaviour. When TF trees, controller timing, or planning-scene synchronization drift, I can trace the fault from the planner down to the bus frame.
03
Manipulation people can use
RGB-D perception, TF-consistent state estimation, and planning-scene management feed environment-aware manipulation behaviours, integrated with teleoperation and operator tooling into a single deployable system. Versioned launch configurations, controller parameter sets, and documentation make bring-up, regression testing, and field deployment repeatable.
R1D1: founding team to CES 2025 in four months
As a founding robotics engineer at a domestic-robotics startup, I owned the software platform for its first mobile manipulator. Four months from first commit, R1D1 executed a live autonomous grasp-and-handover pipeline at CES 2025: detection, 3D pose estimation, grasp planning, and execution on real hardware in an uncontrolled show-floor environment. I drove the systems integration behind it, aligning robot models, interfaces, and test coverage across mechanical, electrical, and software teams in the US and China.
01
The CES demo
A complete perception-to-actuation pipeline running live: object detection, depth-based 3D position estimation, MoveIt grasp planning, and closed-loop handover to a person. The engineering was in the integration and reliability work, getting perception, planning, and control to agree on one robot and hold up across repeated public runs, on hardware that had been a concept four months earlier.
02
VR teleoperation
I designed the real-time streaming architecture for a Meta Quest 3 teleoperation system: onboard RGB-D data serialized into a live point cloud with a 100 to 200 ms end-to-end latency budget covering capture, encoding, transport, and in-headset rendering, with operator commands streamed back to the robot over the same link.
Simulation, control, and the occasional field robot
01
Autonomous power-line inspection
National Research Council Canada · 2026
An NVIDIA Isaac Sim environment for an autonomous power-line inspection system: articulated robot and environment assets with physically meaningful properties, full ROS 2 integration for state, perception, planning and control, and reusable scenarios that test autonomy across poses, viewpoints, and operating conditions before deployment.
02
SO-ARM100 sim-to-real policy learning
Open source · 2026 – present
An open platform for reinforcement learning and vision-language-action experiments: an Isaac Sim digital twin and ROS 2 pipeline for simulated data collection, policy training and evaluation, and deployment of learned manipulation policies, including VLA models like π0, to a physical arm.
03
Four-wheel swerve-drive AMR
Open source · 2024
Derived the inverse kinematics and wrote a custom ROS 2 controller turning joystick input into per-wheel steering and velocity commands for stable omnidirectional motion in Ignition Gazebo. Published openly: 30,000+ views, 600+ engagements, and a repost from Open Robotics.
04
Kisaan Rover
Agricultural UGV · 2022–23
Led an eight-wheel autonomous agricultural UGV for seeding and weeding missions, built on ArduPilot, Jetson Nano, and RTK GPS for navigation and localization. Solved the control problems that come with an eight-wheel suspension, backed by ₹250,000 in government funding.
05
Welding carriage automation
Larsen & Toubro
A motorized welding carriage automating long structural weld runs, with drive control and speed regulation tuned for consistent weld quality on industrial workpieces.
Simulation & modeling
Robotics
Programming & tooling
Mechanical design
Experience
- Robotics Engineer · Sielo Robotics 2025 –
- Founding Robotics Engineer · Open Droids 2024 – 25
Education
- M.Eng, Electrical & Computer Eng. · University of Ottawa 2024 – 26
- B.E., Mechatronics · Gujarat Technological University 2019 – 23
Recognition
- Live robot demo at CES 2025
- Open Robotics reposted the swerve-drive AMR
- Robofest Rover finals · ₹250k development grant
- Student Startup Policy of India · ₹150k grant
- ABU ROBOCON 2022 · zone-best first-stage score
- Published in Electronics For You magazine
Let’s build the next robot.
Email is fastest, and I usually reply the same day.