Rowboat vs Velo

Rowboat is a local-first, open-source AI coworker that connects to your email and meeting notes, builds a persistent Obsidian-compatible knowledge graph from them, and uses that context to draft documents, meeting briefs, slide decks, and emails. It works with local models via Ollama or LM Studio, or with hosted APIs, and supports MCP for connecting external tools. The design philosophy is deliberately anti-cloud: all data stays in plain text Markdown files you can read, grep, and version-control. The knowledge graph is transparent — you can open it in Obsidian and see exactly what the AI knows about you. No black-box embeddings in a proprietary vector store, no "trust us with your emails" data agreements. Rowboat implements what Karpathy described as a "long-term memory coworker" — an AI that compounds value over time because it actually knows your history, your projects, and your terminology. TypeScript codebase, Apache 2.0 license, surging on GitHub trending this week.

Velo lets you record or upload anything — slides, PDFs, docs, screen recordings, websites — and instantly converts it into a polished video message narrated by a hyper-realistic AI avatar with lip sync, eye blinks, and natural gestures. The whole workflow runs in-browser with no downloads required. The key insight is async communication fatigue: teams are drowning in wall-of-text Slack messages and poorly-produced Loom videos, but nobody has time to polish a proper recording. Velo fills the gap by letting you share a PDF, pick a voice, and ship a professional-looking walkthrough in under two minutes. It launched on Product Hunt today and hit #1 with 464 upvotes — unusually strong traction for a non-developer tool. The avatar quality is notably better than earlier AI presenter tools. Early users are reporting it as a replacement for Loom in cases where they want a "polished" look without showing their face or spending time on editing.