AI use cases
551 ways people actually use AI, compiled from the videos in this
library. Each names the job and why it is worth doing, and links to the
moment in the source video.
-
Browser-mediated task execution
— Agentic browsers use AI to interpret requests and perform actions across authenticated browser sessions.
-
Producing software
— Coding agents make it cheaper and faster to build software and may enable more ambitious initial products.
-
Computing or looking up information
— AI agents can perform retrieval and calculations, but direct knowledge remains faster for reasoning.
-
-
Draft a cold email using the user's context files.
— Show how an agent can use stored background, values and business information to produce context-aware writing.
-
Record corrections and preferences in a memory markdown file.
— Allow the agent to improve across sessions and apply learned rules consistently.
-
Build markdown context files by interviewing the user.
— Capture personal, business, brand, offer and customer information as reusable assets for agents.
-
Generate marketing campaign assets and configure campaign drafts across connected services.
— Replace manual copying between project management, commerce, email and advertising systems with end-to-end agent execution.
-
Analyze 60 Instagram reels and create an HTML report covering performance, angles, hooks, topics, titles and why each video worked.
— Demonstrate an agent's observe-think-act loop and its ability to find files, retrieve missing data and produce a finished report.
-
Agent-step verification
— Check each agent action before the next one to improve reliability.
-
Autonomous task execution
— Let a language model decide actions, execute them, observe results, and repeat in an agentic loop.