AI Studio Architecture, Specs & FAQ
Autonomous Local Development • Engineering GuideAlpha Software Notice: JV Logic AI Studio is currently under active engineering and optimization. Features described below reflect the latest build on our private staging branch.
1. Hardware Requirements & Recommendations
| Component | Minimum Spec | Recommended Workstation |
|---|---|---|
| GPU / VRAM | 6 GB VRAM (RTX 3060, 2060) | 8 GB - 16 GB+ VRAM (RTX 4070 / 4080 / 4090) |
| System RAM | 16 GB DDR4/DDR5 | 32 GB+ DDR5 |
| Storage | 10 GB SSD space | Fast NVMe SSD (M.2) |
| Operating System | Windows 10/11 x64 or Ubuntu 22.04+ | Windows 11 with WSL2 (Ubuntu 24.04) |
2. Frequently Asked Questions (FAQ)
Q: Do I need programming experience or complex developer tools to use AI Studio?
No. AI Studio is purpose-built for people at home and coding beginners all the way through to advanced engineers. You do not need Docker, WSL, Android Studio, or Xcode installed to get started. Beginners can simply describe what they want in plain English (e.g. "Create a simple checkers game" or "Build an offline budget calculator"). The AI agent writes the code, self-corrects any errors, and renders the result live in the Web Preview tab. Advanced users can tap into local GPU model tuning (CUDA, ROCm, Vulkan), headless Linux servers over SSH, and VM sandboxes whenever needed.
Q: How does AI Studio run without an internet connection?
AI Studio ships with integrated support for local GGUF quantised models. When running locally, it launches an embedded llama.cpp server that loads models directly into your graphics card VRAM. Everything — token generation, prompt evaluation, tool decision making, file writing, and syntax checking — happens inside your workstation.
Q: How do Universal Apps work on iPhone without a Mac or Apple Developer account?
Universal Apps generated by AI Studio utilize modern Progressive Web App (PWA) standards. They include an offline Service Worker (sw.js), a full web manifest (manifest.json), and notch-safe responsive CSS. When opened on iPhone Safari, users simply tap "Share → Add to Home Screen". The app installs as a standalone fullscreen icon on the iOS home screen with offline caching, high-accuracy GPS tracking, and zero App Store fees.
Q: What hypervisors and sandboxes are supported?
AI Studio automatically scans for WSL2 (Windows Subsystem for Linux), Microsoft Hyper-V, and Oracle VirtualBox. You can also configure base Linux VM profiles (e.g. Debian, Ubuntu, TrueNAS, or Proxmox) via SSH. Once connected, commands executed by the autonomous agent run inside the isolated sandbox rather than directly touching your host system.
Q: Are my company's files or API keys ever uploaded?
Never. AI Studio includes strict regex-based secret redaction and has zero outbound telemetry beacons. If you configure a private in-house Ollama or vLLM server, requests travel strictly over your local area network (LAN).
3. Contact & Technical Inquiries
For alpha feedback, partnership inquiries, or custom private server deployments:
Engineering Desk: [email protected]
Physical Location: Melbourne, Australia