Jacob Furtaw
Crafting intelligent AI systems that push the boundaries of what's possible.
Featured Projects
Selected work from the last few years
Cloak AI 2.0
Cloak AI 2.0 is a fully local, privacy-first AI workspace built for real productivity. It runs on Ollama with seamless dynamic model switching and features reliable tool calling with exceptional transparency — every reasoning step and tool invocation is visible in real time through the Agent Activity sidebar, while sources appear in a dedicated panel. The system includes a proper RAG implementation using ChromaDB with bge-m3 embeddings and named collections that users can create, switch between, and manage directly from the UI. Documents can be added as one-shot context or ingested into persistent knowledge bases. Multimodal input is first-class: users can attach up to 5 images and documents in a single message (with drag & drop support), and the agent can reference previous images across turns using explicit labels. The agent can also search the web for images and display them inline in responses. Additional local capabilities include high-quality document parsing with NVIDIA Nemotron Parse, voice input via Whisper Large v3 Turbo, satellite and general image segmentation with a fine-tuned SAM3 model (integrated with Google Earth Engine), and advanced weather forecasting using Earth2Studio + FuXi with animated GFS visualizations. Everything runs locally with zero data leaving the device.
Chat RAG
Advanced Retrieval-Augmented Generation (RAG) coding assistant with support for multiple LLMs, documents, and GitHub repositories.
Agent Qwen
Multimodal AI agent designed to emulate a Geek Squad Consultation Agent. Built on Qwen2.5-Omni-7B that can understand and respond with text, images, audio, and video. Designed as a Geek Squad-style troubleshooting assistant with voice output.