Who I am & where I'm headed
Final-year Mechatronics Engineering student at Sathyabama Institute of Science & Technology, working at the intersection of AI and robotics.
Hands-on across the full robotics stack — ROS/ROS2, MoveIt, RViz, Gazebo, URDF, and hardware control via Arduino/ESP32 — with deep AI experience spanning LLMs, agentic systems, multimodal inference, computer vision, and reinforcement learning.
Currently COO at Avixr Technologies and an active open-source contributor to LangChain with 64 GitHub stars across 37 repositories and 1,241+ commits.
Primary Focus
Secondary
Sathyabama Institute of Science & Technology
B.E. Mechatronics Engineering (Final Year)
June 2023 – May 2027 · Chennai, India
Bright International School
Higher Secondary Education (CBSE)
2021 – 2023 · Ahmedabad, Gujarat
Delhi Public School
Secondary Education (CBSE)
Mar. 2021 · Gandhinagar, Gujarat
Tools & technologies I build with
Roles, leadership & internships
Avixr Technologies Pvt Ltd · Chennai, India
IEEE IAS – Student Branch Chapter · Chennai, India
Sathyabama Institute of Science & Technology · Chennai, India
Wheels India Limited · Chennai, Tamil Nadu
Filter by Robotics · AI · CS

End-to-end ROS2 control pipeline driving a physical robotic arm in real time using MoveIt for motion planning, bridging ROS2 nodes to Arduino-controlled servos over serial for closed-loop feedback.
Hover to preview
Click to play
🤖 The Next-Gen AI Agent. Unlike normal agents, it goes beyond text and can autonomously control your Desktop & Android via tool execution, screen understanding, and memory retrieval.
Hover to preview
Click to play
A planning-centric architecture for automated educational video generation using LLMs, transforming educational content into structured video scripts and visual assets via multi-stage pipelines.

Vision-guided 6-DOF robotic arm for autonomous pick-and-place, integrating OpenCV object detection with LLM-based high-level task planning across ROS2, Gazebo, and RViz.

An intelligent terminal application using NLP to interpret, explain, and execute shell commands via a conversational AI interface. Evolved across 4 major versions (Python → TypeScript).

ROS2 package mapping RViz joint-state commands to Arduino servo actuation on a SCARA manipulator with URDF-defined joints, tf2 transforms, and a lightweight serial bridge for low-latency command streaming.

Web-based 3D robot visualizer that parses URDF/Xacro and renders articulated robots in the browser via Three.js — a lightweight alternative to RViz with real-time joint updates and dual-camera views.

Virtual digital twin of a SCARA manipulator in RViz with accurate URDF kinematics and joint articulation for validating motion planning and workspace reachability before hardware deployment.
Hover to preview
Click to play
Controlling servos using ROS2 and RViz with Arduino integration for precise robotic motion, with URDF-based robot modeling.

Full robotic-arm simulation on ROS Noetic, modeling links and joints in URDF and validating motion in RViz, demonstrating cross-distro fluency between ROS1 (Noetic) and ROS2.

An interactive 2-DOF robotic arm simulator with real-time inverse kinematics, visualization, and step-by-step mathematical explanations using Gradio.
An AI-powered Snake game using reinforcement learning and neural networks to train an autonomous game-playing agent.
Full-stack AI-powered fitness assistant providing personalized exercise guidance via a conversational AI interface with Google OAuth authentication and customized workout recommendations.
A Python library published on PyPI for seamless serial communication with Arduino boards, featuring auto-detection and multithreading support.
A modern, responsive portfolio website built with Next.js, Tailwind CSS, and Framer Motion.
Proprietary end-to-end multimodal AI inference engine with native Speech-to-Text, Text-to-Speech, and Vision-Language Model components using ONNX Runtime and Apple MLX for optimized on-device inference.
Giving back to the ecosystem
Authored PR #2259 adding the complete Vision + Multimodal section after identifying missing Groq vision support (issue #34589). Now live in official docs, enabling adoption of vision-capable models like Llama-4 Scout/Maverick.
Contributed to the core LangChain framework, improving developer experience for LLM-powered application development.
Research & preprints
Gokulbarath SK · Preprint, Zenodo · Aug. 2025
Open source by the numbers
My open source contributions
Total Stars
0
Repositories
0
Total Commits
0
Pull Requests
0
What I'm working on
🤖 The Next-Gen AI Agent. Unlike normal agents, it goes beyond text and can control your Desktop & Android.
Feel free to reach out for collaborations or just a friendly hello