Summary
Matthew Berman walks through twelve practical ways to use a personal AI assistant, demonstrated mainly in Grokbot and applicable to similar assistants such as Muse and Dot. Each use case comes with a template or a copy-paste prompt, collected in a document hosted on Box.
Examples include a money-saving bot that scans email for subscriptions, drafts negotiation emails to lower prices, and found over $1,000 a month in fixed seats across three Gmail accounts. It also searches California’s unclaimed property site for money owed. Another bot connects to a Tesla through the official fleet API to precondition the cabin, heat the seats, open the garage, and set navigation from the calendar. A weekly cleanup bot on a Mac Studio finds cache and stale files, such as about 50 GB of clearable data, and lists them without deleting anything.
Further use cases include a family bot that reads school and activity emails and adds events to a shared calendar with a daily digest. A lead bot monitors DMs on Instagram, LinkedIn, and X for sponsorship offers, archives spam, and drafts replies for approval. Berman stresses keeping the assistant from acting without the user’s okay on sensitive actions.
📺 Source: Matthew Berman · Published October 07, 2026
🏷️ Format: Tutorial Demo
