What if Slack could take a conversation, figure out what needs to happen next, and get it started?
Slack promised to replace email, but channels, pings, and follow-ups can become another layer of work. For my Late Checkout Agency interview assignment, I explored a different direction: make most conversations unnecessary by giving AI the context and permissions to act.
Twin represents a teammate through shared context and remembers to follow up. Producer turns a conversation into work: bringing in context, asking an agent for help, and turning a decision into action. Reins lets the person decide what AI can do on its own and what needs approval.
The walkthrough follows Janice, an illustrative product lead, through a decision that needs her judgment. The concept moves from surfacing that decision to acting on it and following up, while keeping control with the person.
I developed the product-strategy narrative, interface concepts, animated demonstrations, and presentation. The 14-slide deck and video connect the proposed experience to changing work habits, agents acting as teammates, retrieval, tool use, and agent-to-agent communication.
This is an interview concept, not a commissioned Slack project or a shipped Slack feature. The people, workflow, and weekly savings illustrate the proposal; they are not measured customer results. The accompanying deck and video document the design argument and proposed experience, not a validated business outcome.