Jim and Claude got the travel line in the Q3 budget down 10% this afternoon. The budget balances once Pam sends her numbers, and Friday's summary is the next thing to write. Then he went home, where he uses ChatGPT, and ChatGPT asked him to catch it up.
This is the thirty-five-second version of not doing that.
Watch it, with sound. The lines below are the voiceover, word for word.
What happens
Jim worked through the budget with Claude at work. Tonight he's on ChatGPT at home.
And ChatGPT has no idea where he left off.
RelayLink connects one AI to the next.
Jim tells Claude to save where they left off. RelayLink keeps it, just for him.
At home, he asks ChatGPT to pick up where he left off.
It knows the numbers and the next step, so he keeps going. Nothing to re-explain.
What Jim says, leaving
Save where we left off in RelayLink.
Claude writes it down in one go: where the work stands (travel cut 10%), what it is waiting on (Pam's numbers) and the next step (draft the Friday summary). There is no note to approve, because a handoff goes to nobody but Jim. It is saved, and only he can see it.
What Jim says, arriving
Pick up where I left off in RelayLink.
ChatGPT reads the handoff and carries on from the next step: travel is cut 10%, Pam's numbers are outstanding, Friday's summary is next, and would he like it drafted now. Yes. Nothing was pasted and nothing was re-explained.
It is not only work and home
The same move takes a plan from a chat into Claude Code, a morning's research from ChatGPT into an afternoon in Claude, or today's thread into tomorrow's fresh one in the same app. Any AI connected to your account can save a handoff, and any of them can pick the latest one up.
Try it
Connect RelayLink to the AIs you use and say the first sentence above the next time you stop. The handoffs page has what a handoff carries and what it does not.