My Independent AI 🤖

A fully local, privacy-first personal AI assistant. Ask questions about your own data — emails, documents, chats — without anything leaving your machine.

Stack: Ollama · Qdrant · Streamlit · Python (uv monorepo)

---

Quickstart

Prerequisites: Docker & Docker Compose

git clone https://github.com/your-username/myindependent-ai.git
cd myindependent-ai
docker compose up

Open http://localhost:8501 — that's it.

First run: docker compose up automatically pulls llama3.2 (~2 GB) and nomic-embed-text (~300 MB) via the ollama-init service. Subsequent starts are instant.

---

What's Running

Service	URL	Purpose
Dashboard	http://localhost:8501	Streamlit chat UI
Ollama	http://localhost:11434	Local LLM + embeddings
Qdrant	http://localhost:6333	Local vector database

---

Getting Data In

The dashboard searches whatever is in your Qdrant personal_data collection. To populate it, run one of the importers:

# Gmail (requires OAuth credentials)
uv run python -m importers.orchestrator --importer gmail

# WhatsApp exports
uv run python -m importers.orchestrator --importer whatsapp

# Files from Synology NAS
uv run python -m importers.orchestrator --importer synology-nas

See docs/CONTRIBUTING.md for adding new importers or setting up your dev environment.

---

Local Development (without Docker)

# Install all dependencies
uv sync --all-packages --group dev

# Run the dashboard
uv run streamlit run apps/admin-dashboard/app.py

# Run tests
uv run pytest

You'll need Ollama and Qdrant running separately:

ollama serve                  # terminal 1
docker run -p 6333:6333 qdrant/qdrant   # terminal 2

---

Configuration

Copy .env.example to .env and adjust if needed:

cp .env.example .env

Key variables:

Variable	Default	Description
OLLAMA_BASE_URL	http://ollama:11434	Ollama API URL
QDRANT_URL	http://qdrant:6333	Qdrant URL
MAPPING_DB_PATH	/data/mapping.db	PII mapping database path

---

Project Structure

.
├── apps/
│   ├── admin-dashboard/    # Streamlit chat UI
│   ├── importers/          # Data ingestion (Gmail, WhatsApp, etc.)
│   └── orchestrator/       # Ingestion pipeline runner
├── libs/
│   ├── embedding/          # Ollama embedding wrapper
│   ├── vector-storage/     # Qdrant client wrapper
│   └── privacy-core/       # PII scrubbing (Presidio + SQLite)
├── infrastructure/         # Terraform for GCP deployment
├── scripts/                # Operational helpers
├── docker-compose.yml      # Local stack (Ollama + Qdrant + Dashboard)
└── pyproject.toml          # uv workspace root

---

Architecture

See docs/architecture.md for the full system design. For GCP/cloud deployment, see infrastructure/.

Upcoming Architectural Milestones

See Technical Roadmap for our technical roadmap.