rowboat: what it is, what problem it solves & why it's gaining traction

What it solves

Rowboat is a local-first AI coworker designed to eliminate the need to repeatedly re-explain context to AI. It solves the problem of "cold start" retrieval by maintaining a long-lived, compounding knowledge graph of your work—including people, projects, and decisions—rather than just searching transcripts or documents on demand.

How it works

Rowboat connects to your email (Gmail), calendar, and meeting notes (Fireflies or native notes) to build a memory system. This memory is stored as an Obsidian-compatible vault of plain Markdown notes with backlinks on your local machine. It uses this structured context to perform tasks like drafting emails, preparing meeting briefs, and generating PDF slides. Users can bring their own LLM (via Ollama, LM Studio, or hosted APIs) and extend the system's capabilities using the Model Context Protocol (MCP) to connect to external tools like Slack, GitHub, and Jira.

Who it’s for

It is built for professionals who need an AI assistant that remembers their specific work history and commitments across different communication channels while maintaining full local control and privacy of their data.

Highlights

• Local-first Memory: All data is stored as plain Markdown files on your machine, making it inspectable and editable.
• Knowledge Graph: Maintains explicit relationships between entities to create compounding context over time.
• Live Notes: Automatically updates notes on specific topics, people, or competitors using the @rowboat command.
• Extensible Tooling: Supports MCP servers and Composio tools for integration with a wide range of external services.
• Model Agnostic: Compatible with both local models (Ollama, LM Studio) and hosted API providers.