AI use cases

2093 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.

Let a teammate join and contribute to a live AI coding-agent session from another terminal.

ManuAGI - AutoGPT Tutorials
Run a long-lived, always-on AI agent in the cloud.

ManuAGI - AutoGPT Tutorials
Rewrite Claude Code responses into plainer language while preserving the agent and saved transcript.

Github Awesome
Review and edit an HTML or Markdown document before sending the complete set of human feedback to an agent.

Github Awesome
Convert a technical book or documentation folder into a compact agent skill that loads only the material relevant to a question.

Github Awesome
Rewrite aerospace instructions so one AI agent can pass them to another without clarification.

Github Awesome
Control whether an AI agent's proposed actions are allowed, escalated, or blocked before execution.

Github Awesome
Give coding agents persistent, searchable project memory across sessions.

Github Awesome
Dictate code or instructions directly into the Pi coding agent.

Github Awesome
Diagnose PostgreSQL health and provide database evidence to scripts or AI agents.

Github Awesome
Check an agent's next step before it commits to it.

IBM Technology
Use a CLAUDE.md or agents.md file to preload an agent with stable information about a person and business.

Open Residency
Aggregate many application connections behind one connector so the same tools are available across agent harnesses.

Open Residency
Let AI agents read, edit, render, validate, and automate Word, Excel, and PowerPoint files.

ManuAGI - AutoGPT Tutorials
Run agent-like workloads at higher density and lower latency on Kubernetes.

ManuAGI - AutoGPT Tutorials
Connect AI agents to team communication and development channels so people can tag and use them as teammates.

ManuAGI - AutoGPT Tutorials
Define enterprise AI agents as version-controlled files with explicit context, skills, tools, permissions, and identity.

ManuAGI - AutoGPT Tutorials
Give multiple AI agents access to MCP servers through one local gateway while reducing tool-context overhead.

ManuAGI - AutoGPT Tutorials
Let AI agents perform web tasks inside logged-in browser accounts.

ManuAGI - AutoGPT Tutorials
Build an enterprise-ready AI agent from customer transcripts and documentation.

a16z
Create, test, deploy, and continuously improve an enterprise conversational agent.

a16z
Scaffold a new spec-driven development project with agent-specific prompts, templates, helper scripts, and project metadata.

Den Delimarsky
Create an open-ended AI coding assistant in which agents plan a solution, write code, review it, recommend changes, and help debug it.

IBM Technology
Control what an AI agent can do with a user's email.

IBM Technology
Implement time-limited, transaction-specific permissions for agent actions.

IBM Technology
Use a parent agent to decompose a larger objective into several child coding sessions.

Greg Isenberg
Manage a multi-step coding objective by delegating work to child agents.

Greg Isenberg
Use agents to index and cross-reference medical information for a child with a rare form of epilepsy.

Y Combinator
Create a complete marketing package for an individual real-estate listing so it can be offered to the listing's agent as a sales sample.

Chris Koerner on The Koerner Office Podcast
Apply policy-based access control when connecting agents to enterprise systems.

IBM Technology
Use interaction telemetry to remove or narrow agent and user permissions.

IBM Technology
Automatically create and validate prompts used by agents.

IBM Technology
Create a design frame from a natural-language request through an agent connected to Paper.

Y Combinator
Enforce the TDD cycle for a generative-AI coding agent without repeatedly reminding it to follow the process.

Modern Software Engineering
Create separate working versions of an application so agents can develop different ideas concurrently.

Riley Brown
Send data from a custom app into Buzz each day so agents can analyze and act on it.

Riley Brown
Transfer an existing set of skills and memory into a focused management agent.

Riley Brown
Create a Codex-powered research agent named Harry and a public research channel.

Riley Brown
Lead a research-channel discussion to structure automation for finding ideas for the Agent Native podcast.

Riley Brown
Research and design the technical infrastructure needed to host multiple AI agents safely.

Chris Koerner on The Koerner Office Podcast
Convert audience interest into a hosted AI-agent community with recurring subscriptions.

Chris Koerner on The Koerner Office Podcast
Generate editable Astro documentation sites with layouts, components, styles, routes, agent-readable markdown twins, llms.txt, and JSON-LD.

Github Awesome
Monitor coding-agent activity locally and generate live alerts or forensic timelines.

Github Awesome
Host several coding agents on the same repository while preserving their native terminal interfaces and shared context.

Github Awesome
Provide scalable isolated environments for large fleets of agents.

Github Awesome
Distill books, videos, podcasts, interviews, and courses into callable agent skills.

ManuAGI - AutoGPT Tutorials
Run an AI coding agent inside a disposable, network-restricted Linux virtual machine.

ManuAGI - AutoGPT Tutorials
Pass agent handoff artifacts between subagents without copying the underlying file.

ManuAGI - AutoGPT Tutorials
Replay a coding agent's session as a visual map of the files it searched, read, and edited.

ManuAGI - AutoGPT Tutorials
Control multiple coding agents and trigger coding workflows from a physical command surface.

ManuAGI - AutoGPT Tutorials