const ayuugoyal = ( name, passion ) =>

Hi, I'm Ayush 👋
Physical AI & Robotics Engineer with a passion for building robots that think.
AG

About

I'm Ayush Goyal — an AI Engineer focused on Physical AI: bringing intelligence into the real world through robots and embedded systems. I build production-grade GenAI systems — RAG pipelines, LLM integrations, MCP servers — and the robotics software that puts them into motion, from ROS2 control stacks and Arduino firmware to patented robotic hardware. Currently an AI (Harness) Engineer at TAP Innovations (remote, US), I'm betting on Physical AI as the next big thing and building toward it. I ship things that work in the real world — enterprise chatbots with paying customers, open-source AI platforms, and robots you can actually control.

Work Experience

TAP Innovations

May 2026 - Present
AI (Harness) Engineer
AI (Harness) Engineer at TAP Innovations, a US-based software startup building integrations, analytics, and custom applications for enterprise systems. Designing and shipping AI-driven solutions in a fully remote role, while sharpening my focus on Physical AI — bringing machine intelligence into real-world robots and embedded hardware, which I believe is the next big thing.

Loadshare Networks

Nov 2025 - Apr 2026
AI Engineer
Worked as an AI Engineer at Loadshare Networks, a leading logistics and supply chain technology company. Developed and implemented AI-driven solutions to optimize logistics operations, enhance delivery efficiency, and improve customer experience. Additionally provided freelance consulting services for AI/ML projects across various domains.

Data Alt Dynamics

July 2025 - Nov 2025
Backend Developer
Deployed a Raspberry Pi–based server with multi-sensor integration for real-time data acquisition and control. Designed and optimised local ComfyUI workflows, boosting AI-driven process efficiency by 70%. Built scalable backend architectures and automated pipelines by integrating Cloudflare Tunnel with n8n, enabling secure remote access, seamless cloud sync, and reducing manual operational tasks by 40%.

Physics Wallah

May 2025 - July 2025
Trainee
Contributed to backend development and platform engineering at India's leading ed-tech platform serving millions of students across JEE, NEET, and school curricula.

Bharat Electronics Limited

June 2024 - Aug 2024
Summer Intern
Worked on embedded systems and electronics projects at India's premier defence electronics manufacturer under the Ministry of Defence, gaining hands-on experience with industrial hardware systems.

Open Source Contributions

  • a

    archestra-ai/archestra

    AI agent platform — MCP servers, multi-LLM orchestration, RAG knowledge connectors & embeddings. Contributed Notion connector, Gemini embeddings, SharePoint & OneDrive connectors, OAuth proxy fix, and recursive subfolder traversal across 6 PRs. Awarded $400 in bounties by maintainers.
  • b

    bolna-ai/bolna

    Open-source voice AI agent framework (telephony, LLM, TTS pipelines). Resolved duplicate .env file requirement in telephony server Dockerfiles (Twilio & Plivo), centralising config via docker-compose env_file directive (PR#310).

Patents

  • 3

    3D Printed, Pneumatically Actuated Four-Finger Adaptive Gripper

    UK Design Registration for an innovative industrial robotic end-effector featuring pneumatically actuated multi-finger grippers capable of grasping objects of various shapes and sizes. This patented design represents a breakthrough in adaptive robotics technology for manufacturing applications.

Skills

Physical AI / Robotics

ROS2
ROSBridge
Arduino
Raspberry Pi
Embedded Systems
Motion Control
Sensor Integration
Computer Vision
WebSockets

AI / GenAI

LangChain
LangGraph
LangSmith
OpenAI API
Gemini API
HuggingFace
RAG
VectorDBs
MCP Servers
Prompt Engineering
LLM Fine-tuning

AI Dev Tools

Claude Code
Cursor
Antigravity
n8n

Languages

Python
TypeScript / JavaScript
C++

Backend & Frameworks

Next.js
FastAPI
Flask
Express.js
React.js
GraphQL
REST APIs

Databases & Infrastructure

PostgreSQL
MongoDB
MySQL
Docker
AWS / Azure
Nginx
PrismaORM
DrizzleORM

Projects

Physical AI / Robotics

    BCN3D Moveo Control System

    BCN3D Moveo Control System

    A complete control system for the BCN3D Moveo robot arm with a modern Next.js web interface, ROS2 integration, and Arduino firmware. Features include real-time WebSocket communication, joint and cartesian control modes, 3D visualization, and system health monitoring.
    Next.js
    TypeScript
    ROS2 Humble
    ROSBridge
    Python
    C++
    Arduino
    pnpm
    WebSockets
    SCARA Robot

    SCARA Robot

    A comprehensive control system for a 4-axis SCARA robot that combines an Express.js backend API, Next.js frontend, and Arduino Mega firmware for precise motion control, enabling remote operation and automation via a web interface.
    Arduino Mega
    Nema 17 Stepper Motors
    Next.js
    Typescript
    TailwindCSS
    Shadcn UI

AI

    Chatter AI

    Chatter AI

    Chatter AI is a lightweight, embeddable AI chatbot built to help businesses provide instant, intelligent customer support. It answers product queries, recommends related items, and integrates in under 10 seconds.
    Next.js
    Typescript
    PostgresSQL
    DrizzleORM
    Tailwindcss/ShadcnUI
    Google Gemini API
    Clerk
    LangChain
    LawGPT

    LawGPT

    LawGPT is an AI-powered legal assistant that uses Retrieval-Augmented Generation (RAG) to provide instant, document-backed legal guidance. By leveraging advanced technologies like HuggingFace embeddings, FAISS vectorstore, and the Gemini API, it helps users navigate complex legal matters efficiently and accurately.
    Next.js
    Typescript
    NeonDB - PostgresSQL
    DrizzleORM
    Tailwindcss/ShadcnUI
    Google Gemini API
    HuggingFace Embeddings
    FAISS Vectorstore
    Clerk
    FastAPI
    LangChain
    QuickDocs

    QuickDocs

    Instant documentation fetcher for any service. CLI + MCP server powered by llms.txt — fetch docs for any library or API in seconds directly from your terminal or AI agent.
    TypeScript
    Shell
    MCP Server
    CLI
    llms.txt
    Node.js

Get in Touch

Got something on your mind? Slide into my DMs on Twitter with a direct question or mail at ayushgoyal8178@gmail.com, and I'll hit you back when I can.

—visitors