> INITIALIZING...

> LOADING PROFILE...

> READY

SYSTEM ONLINE // SALINAS, CA

ROBOTICS
ENGINEER

Embodied AI · Robot Learning · Safe Autonomy

> I build learning-enabled robotic systems that perceive, plan, and act safely in the physical world.

> Robotics Engineering Intern @ AgRobotics

> Master's in Autonomy & Robotics @ UIUC, graduating Dec 2026

01

FEATURED_WORK

Robotics Engineer, Bi-Manual Harvesting Robot

End-to-end autonomy for an agricultural robot at AgRobotics: a fine-tuned pi0.7 VLA policy for harvesting, with dual-arm planning, berry perception, and in-field navigation.

VLA trained on 400+ hrs of teleop data Real-time collision-aware dual-arm planning

Research Engineer, Robotic Microsurgery

Real-time impedance force control for a KUKA LBR Med 7 + hexapod, part of an ARPA-H-funded AI-guided microsurgery project at UIUC.

<10ms command latency 10µm 3D reconstruction (tissue phantoms)

O-VLA: Trajectory Optimization Safety Layer

A morphology-agnostic optimizer that turns unsafe VLA-generated trajectories into collision-free, constraint-satisfying motion.

100% constraint satisfaction 5 morphologies, 6-23 DOF

Vision-Language-Action System for Manipulation

An end-to-end pipeline turning unseen YouTube cooking videos into executable robot commands, trained on 1.6k+ clips.

77-85% action accuracy (held-out clips) 85-90% success in AI2-THOR

Cross-Embodiment Robot Perception

A cross-attention ViT conditioned on kinematic features, producing representations that transfer to unseen humanoid robots.

99.3-100% classification vs. 20% baseline

Humanoid Whole-Body Motion Planning

ZMP preview control, A* footstep planning, and MPC balance optimization for the Unitree G1, with RL-based locomotion.

49% lower energy vs. PD baseline 75%+ task success (MuJoCo)
02

MISSION_LOGS

RUNTIME_HISTORY (WORK)

Aug 2026 - Present

Robotics Engineering Intern

AgRobotics, Salinas, CA
Project: Manipulation and Navigation for Agriculture Robots
  • Deploying a fine-tuned pi0.7 VLA policy on a bi-manual harvesting robot, trained on 400+ hours of teleop data.
  • Implemented dual-arm, collision-aware motion planning with Pinocchio and RRT, avoiding inter-arm collisions in real time.
  • Built a DAgger pipeline for policy improvement, collecting teleop demos via GELLO in Parquet format.
  • Built an autonomous navigation stack using the ZED SDK for in-furrow tracking, row-switching, and U-turns on farm terrain.
  • Built a berry perception module with a spatial dictionary and pair selector, integrating fine-tuned YOLO detection.
  • Created a simulation environment in Isaac Sim & Gazebo replicating the full robot and autonomy stack for testing.
April 2026 - Aug 2026

Research Engineer

IHSI, University of Illinois, IL
Project: ARPA-H MarginDx, AI-Guided Robotic Microsurgery for Tumor Margin Assessment
  • Built a real-time impedance force-control stack for a KUKA LBR Med 7 + PI hexapod, achieving stable tissue contact regulation.
  • Designed a multi-layer IPC architecture (Java/Python/UDP) bridging the KUKA arm, PC backend, and hexapod with <10ms command latency.
  • Achieved 10µm-resolution 3D tissue surface reconstruction with stable contact force during controlled raster scans on tissue phantoms.
May 2024 - May 2025

Embedded Engineering Intern

Dimension Six Technologies, Mumbai
Project: STM32-based E-Bike Conversion Kit
  • Engineered production-ready firmware with battery management and regenerative braking, extending vehicle range by 40%.
  • Designed a 4-layer PCB in KiCad for a custom ESC, optimizing power stage layout to reduce losses by 15% under peak load.
  • Deployed an IoT stack on ESP32S3 with RFID authentication, enabling remote monitoring and contactless payments over MQTT.
Jan 2024 - Jun 2024

Robotics Research Intern

IIT Bombay
Project: Autonomous Rugged Robot for Military Applications
  • Built a SLAM-based autonomous navigation stack with path planning, achieving 95% accuracy in dynamic indoor environments.
  • Fused IMU, GPS, and RGB-D camera data via an Extended Kalman Filter, improving localization accuracy by 20%.
  • Deployed YOLOv3 for real-time human detection and designed an adaptive gait controller for robust stair-climbing traversal.

KERNEL_VERSIONS (EDUCATION)

Aug 2025 - Dec 2026

Master of Engineering in Autonomy and Robotics

University of Illinois Urbana-Champaign
  • GPA: 3.9/4.0
  • Coursework: Humanoid Robots, Robot Planning, Control & Dynamics, Autonomous Systems, Computer Vision, Deep Learning
  • Key Lab Work: Working with humanoids, manipulators and GEM autonomous platforms.
Dec 2021 - Jul 2025

B.Tech. in Electronics & Telecommunication

Sardar Patel Institute of Technology
  • Specialization in Embedded Systems and Robotics.
  • Minors in Management (S.P. Jain Institute).
03

SYSTEM_SPECIFICATIONS

Where I currently go deepest:

Robot Learning (VLA / IL / RL) Motion Planning & Control Perception & Sensor Fusion Simulation & Sim-to-Real Manipulation

AI_&_ROBOT_LEARNING

Vision-Language-Action Models Vision-Language Models Transformers / Vision Transformers Cross-Attention Contrastive Learning Reinforcement Learning (PPO) Imitation Learning Diffusion Policy / ACT DAgger pi0 VLA Fine-Tuning Foundation Models PyTorch / TensorFlow

AUTONOMY_&_CONTROL

Motion Planning / Path Planning Dual-Arm Collision-Aware Planning (RRT) Teleoperation (GELLO) Capture-Point Balance Control Trajectory Optimization MPC / ZMP Control Control Barrier Functions Force / Impedance Control Jacobian IK / Inverse Dynamics A* / Stanley Control QP Solvers (ProxQP)

PERCEPTION_&_STATE_ESTIMATION

Computer Vision (YOLO, SAM) 6D Pose Estimation ZED SDK / Stereo Depth LiDAR / Point Cloud Processing LiDAR-Camera Fusion BEV Perception SLAM / VIO EKF / Sensor Fusion Localization / Mapping

SIMULATION_&_SOFTWARE

ROS 2 / MoveIt / Nav2 MuJoCo Isaac Sim / Isaac Gym Gazebo PyBullet AI2-THOR Pinocchio ONNX / onnxruntime-web MuJoCo WASM / three.js Linux / Git Real-Time Systems Sim-to-Real / Hardware Validation

EMBEDDED_&_HARDWARE

Python / C++ / MATLAB / Java / Bash STM32 / ESP32 Firmware PCB Design (KiCad) I2C / SPI / UART / CAN MQTT / IoT Fleet Management / OTA Updates Battery Management / Regenerative Braking
04

PROJECT_MODULES

Everything I've built, spanning robot learning, manipulation, autonomous vehicles, and embedded systems.

MuJoCo Web Demo: Unitree G1 Balance

In-Browser MuJoCo Robotics Demos

MuJoCo WASM onnxruntime-web three.js

MuJoCo running fully in the browser: a Unitree G1 you can shove while a capture-point balance controller keeps it standing (250N forward push vs. 50N for bare servos), and a Franka pick-and-place driven live by the diffusion-pick ONNX policy.

VLM Navigation Demo

VLM-Based Autonomous Navigation

GPT-4o ROS2 A* Planning

Intent-based navigation for GEM e4 using GPT-4o for goal localization, BEV perception, and Stanley control. 88% accuracy on KITTI for language-to-goal localization.

6D Pose Estimation Demo

Open-Vocabulary 6D Pose Estimation

FoundationPose SAM-3 InstantMesh

Zero-shot 6D pose pipeline combining VLM semantic inventory, SAM-3 segmentation, and Objaverse mesh retrieval. 76.5-100% ADD-S AUC on YCB-Video, plus an RGB-D variant with ensemble mesh selection.

AutoShield Demo

Pedestrian-Aware AV Safety System

YOLOv11 LiDAR

Real-time pedestrian safety stack fusing LiDAR clustering and RGB-D detection, with Time-to-Collision estimation for proactive speed adaptation. 90% success across tested pedestrian scenarios.

Dynamic Handover Demo

Dynamic Object Handover System

MuJoCo MediaPipe

Vision-based handover with real-time MediaPipe hand tracking, Jacobian IK, velocity limiting, and command smoothing for stable object transport and adaptive release.

RL Locomotion + Safety Layer

PPO Control Barrier Functions

Terrain-adaptive locomotion trained with PPO, filtered by a real-time CBF safety layer. Zero-fall locomotion, 99% unsafe-action rejection, 90% of baseline traversal speed retained.

Terrain-Aware Navigation

Elevation Mapping ROS2

Perception pipeline converting depth clouds to elevation maps for quadruped footstep placement.

Quadruped Locomotion (RL)

PPO Isaac Gym

ANYmal quadruped policies trained via Proximal Policy Optimization, with robust traversal on irregular terrain under zero-fall constraints.

VIO + Footstep Planning

VINS-Fusion Nav2

Fused Visual-Inertial Odometry with footstep planning to enable navigation in GPS-denied environments with <10cm drift.

EMG Prosthetic Arm Demo

EMG-Controlled Prosthetic Arm

ESP32 TinyML XGBoost

A 5-DOF prosthetic arm controlled via EMG signals, with real-time gesture classification on-device at 96% accuracy.

05

VISION_PROTOCOL

vision.md

Robots that learn tasks from watching humans, then execute them safely in the real world.


> # WHAT I'M EXPLORING

How learned behaviors can transfer across robot morphologies while staying physically feasible and safe to execute, combining cross-embodiment perception with trajectory optimization safety layers.


> # ABOUT_ME

I'm a robotics engineer working across robot learning, perception, planning, and control. I like building systems that connect modern AI with the physical realities of robotics, from learned representations and VLA models to real-time force control on actual hardware.

06

ESTABLISH_UPLINK

Always up for talking robotics, embodied AI, and hard autonomy problems.

EMAIL LINKEDIN GITHUB