AI Engineering Studio

Your gaming GPU is
also your AI engine.

We build GPU-powered software that runs on hardware you already own. Working products in active testing -- 20+ repos, 30+ Docker services, real infrastructure. No cloud lock-in, no subscriptions required.

jefe@studio ~ adventure-gui
$ adventure-gui --gpu rtx4070 --campaign "lost-mines"
Loading AI narrator... GPU VRAM allocated (8GB)
Scene illustrator ready · Voice actor loaded
DM> You enter the cavern. Roll perception.
[Scene generated: 1.2s] [Voice: 0.4s] [GPU: RTX 4070]

Product Ecosystem

Working products in active testing. Own the stack, run it locally, encrypt everything.

Active Testing

Adventure GUI

Interactive D&D and tabletop campaign platform with AI narration, real-time scene illustration, and voice acting. Your gaming GPU becomes your AI dungeon master. Runs on an RTX 4060 or better.

AI Narration Scene Gen TTS Voice GPU-Local
Flagship product — alpha testing →
Active Testing

Encrypted Communications

End-to-end encrypted chat with AES-256-GCM, voice/video calls, and cross-platform clients. Web, desktop (Tauri), and Android. 41 crypto tests passing. Deployed and running.

FreeChat FreeVox E2E Crypto LiveKit
Request access →
Active Testing

Personal Wellness Companion

Privacy-first wellness exploration tool. Mood tracking, journaling, and AI-guided reflection -- all running locally. Your data never leaves your machine. Not a medical device; a personal research tool.

JefeHealth Local AI Encrypted Private
Privacy-first — in testing →
In Development

Smart Home + Assistant

Voice-driven personal assistant with environmental monitoring, home automation, and local AI inference. Wake word, STT, TTS.

JefeHome Gigi Whisper Kokoro

Runs on Hardware You Already Own

Our products are designed for consumer NVIDIA GPUs. An RTX 4060 ($300) or RTX 4070 ($550) is all you need. Prefer managed? We offer a hosted tier too.

NVIDIA GPU Live

Development & Hosted Tier

Enterprise GPU for development & hosted services · Consumer GPUs (RTX 4060+) for local use

Target Hardware RTX 4060 / 4070 ($300–$550)
Models Qwen3, DeepSeek-R1, Flux.1
Services 30+ containerized across 20+ repos
Stack Ollama · ComfyUI · Docker · Jenkins CI/CD
Consumer GPU-First

Products are optimized for GPUs gamers already own. An RTX 4060 with 8GB VRAM runs our AI narration, scene generation, and voice synthesis. No enterprise hardware required.

Privacy

Every query, every generation, every conversation stays on your machine. Your data never leaves. No cloud telemetry, no training on your inputs.

Not Vaporware

20+ repositories, 30+ Docker services, Jenkins CI/CD, automated testing. These are working products in active testing, not pitch decks.

Want to try these products?

Chat with AI personalities, explore encrypted communication demos, and see our products in action -- all at jefe.works.

Visit jefe.works

Ask the AI

This is a live demo of our AI platform. RAG-powered, locally hosted, running on our own hardware right now. Want something more playful? Try our 8 AI personality chatbots at jefe.works.

JefeWorks AI
Qwen3 32B + ChromaDB RAG

I'm the JefeWorks AI assistant, powered by local LLM inference and RAG semantic search. Ask me about our products, technology, or AI capabilities.

Engineering Notes

Build logs from the JefeWorks ecosystem -- infrastructure, AI, security, and everything in between.

View all entries →

Get in Touch

JefeWorks is an AI engineering studio based in Illinois. We build GPU-powered software products -- interactive AI experiences, encrypted communication, and personal wellness tools.

jeff@jefeworks.com
Entity JefeWorks LLC
Location Illinois
Founded 2026
Focus GPU-Powered AI Products
JefeWorks AI

I'm the JefeWorks AI assistant. Ask me anything about our products or technology.