Build an AI Assistant That Runs Privately: On-Device LLM Tool-Calling Automation Hub with Quantized Phi-2 on Arduino UNO Q

Every part needed, pre-tested for compatibility, with an AI build companion trained on this exact project. Shipped from Bengaluru in 3-5 days.

Imagine telling your home assistant “If the temperature rises above 30°C and someone enters the room, turn on the fan and log the event” — and having it executed instantly, without any cloud dependency or privacy concerns. This kit makes that a reality by letting you deploy a quantized Phi-2 large language model directly on an embedded Linux MPU, paired with an Arduino UNO Q microcontroller that handles all sensor and actuator interactions through a Visual Bridge tool-calling framework.

What You'll Build

You’ll assemble a fully functional offline automation hub that accepts natural language instructions and maps them to real-world actions. The LLM runs entirely on the included Arduino UNO Q’s Linux subsystem (no cloud required). It interprets your typed or spoken commands, selects the appropriate function from a library of tool definitions — sensor reading, relay switching, RFID access logging, OLED display — and dispatches the execution to the Arduino’s MCU via a serial Visual Bridge. The result is a private, responsive AI that can automate everything from environmental monitoring to access control, all housed in a custom enclosure.

What You'll Learn

• Deploying quantized LLMs like Phi-2 on resource-constrained Linux MPUs
• Designing a Visual Bridge layer to convert natural language to structured function calls
• Interfacing an embedded Linux MPU with an Arduino microcontroller over serial protocols
• Integrating multiple sensor/actuator modules (DHT22, PIR, RFID, relay, OLED) into a tool-calling pipeline

Kit Contents

Why Buy This Kit Instead of Sourcing Parts Separately

Who This Kit Is For

This kit targets advanced makers, B.Tech ECE/EEE students, and professionals prototyping on-device AI solutions. It’s ideal for Smart India Hackathon teams, IIT/NIT/VIT/BITS final-year projects, and ATL Tinkering Labs that want to demonstrate private, natural-language-driven automation without cloud dependence. If you’ve ever wanted to run an LLM locally on custom hardware and control real-world devices through intelligent tool-calling, this kit gives you a complete sandbox.

Built and Backed by Compoden

Every Compoden kit ships with an AI build companion trained on this exact project — accessible via a QR code on the box, with WhatsApp and email backup. We've spent 10 years building projects for makers, schools, and institutions across India. If a part fails because of a manufacturing defect, replace it free within 7 days.

Choose your assembly option:

• Soldering Kit — 25W soldering iron, 60/40 solder wire, flux, and small perfboard for permanent assembly.
• Breadboard Combo — 800-point full-size breadboard with 65-piece jumper wire pack for solderless prototyping.