Jo is the voice assistant that runs your whole stack while you walk and drive. No typing & no screen.
Summarize unread from investors
3 need replies — Chen, Segal…
Reply to Josh: shipping tonight
Sent to #eng-core
Close the WebSocket PR
JO-214 → Done
Move standup to 3:00
Rescheduled — 4 invited
Jo
to you · 0.8s
You —“Jo, close the WebSocket PR.”
Jo —“Found JO-214 — WebSocket integration. Mark it Done?”
You —“Yes.”
Jo, talk to me — 01 / 04
The problem
You open Gmail to check on Sarah’s email. You tab to Slack to answer Josh. You switch to Linear to mark a PR closed. Three apps. Forty-five seconds. You lost the thought you were mid-sentence on.
The problem isn’t the tools. It’s the interface. Keyboards and taps were built for sitting at a desk. Jo is built for the walk to the meeting, the cab, the corner — anywhere you’d rather just say the thing and move on.
“Think Superhuman’s speed obsession, but the interface is voice and the surface area is every connected service — not just email.”
Andrew Chen — a16z
How it works
Jo turns your stack into a conversation, so you can keep moving.
Tap once. Jo listens — hands-free, mid-stride.
It routes to the right tool and calls it. No pipeline you maintain.
You —“Close the WebSocket PR.”
Jo —“Done. JO-214 marked complete.”
You hear the result. You never stopped walking.
Calendar
Standup moved to 3:00 — 4 invited
02 / 04
Integrations
Jo doesn’t rebuild your integrations — it reuses your existing MCP server connections and attaches them directly to the voice session. Every tool you already have access to becomes a spoken command away.
03 / 04
Our philosophy
The slowest thing in your workflow isn’t your Wi-Fi. It’s the app-switching. The wrong tab open. The thought interrupted by a notification you had to tap through to dismiss. We built Jo around a single belief: the interface should disappear. Your tools should answer to you — not the other way around. Speed isn’t a feature. It’s the whole point.
The film
Questions