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.

Run multiple coding agents in parallel — Coordinate agents such as Claude Code, Codex, and Cursor in separate git work trees.

ManuAGI - AutoGPT Tutorials
Embed agentic workflows into applications — Let Copilot handle planning, tool calls, and file edits inside an application.

ManuAGI - AutoGPT Tutorials
Improving models from agent traces — Use routing policies, world model simulations, harness optimization, and distillation to continuously improve an endpoint.

ManuAGI - AutoGPT Tutorials
Selecting tools for an AI agent context window — Expose only the handful of registered or MCP-provided tools that matter for each turn.

ManuAGI - AutoGPT Tutorials
Building AI agents and RAG applications — Provide ready-to-run, provider-agnostic templates for multi-agent teams, voice agents, MCP agents, and memory.

ManuAGI - AutoGPT Tutorials
Providing front-end design guidance — Improve AI coding agents' interface design using commands and deterministic rules.

ManuAGI - AutoGPT Tutorials
Turning books and document sets into agent skills — Let agents answer from locally extracted frameworks, patterns, and chapter content.

ManuAGI - AutoGPT Tutorials
Sharing and reusing AI workflows, skills, MCPs, and connected services — Use the same workflows across agents, tools, teammates, and machines.

ManuAGI - AutoGPT Tutorials
Structuring and optimizing information passed to an LLM or AI agent — Produce more accurate and relevant outputs

IBM Technology
Multimodal agent development — Compose agents, tools, multi-agent teams, knowledge graphs, prompt pipelines, and image, audio, and video generation.

ManuAGI - AutoGPT Tutorials
Web data extraction — Search, scrape, crawl, interact with, and convert websites into formats suitable for AI agents and applications.

ManuAGI - AutoGPT Tutorials
Personalized tutoring — Generate quizzes, conduct research, solve problems with multiple agents, create math animations, and produce interactive visuals from study materials.

ManuAGI - AutoGPT Tutorials
Computer and browser use — Expand agents' ability to test software and operate in a more complete quality-assurance environment.

Y Combinator
Stress testing and code review with sub-agents — Break projects into features, test or review each feature, and direct later testing based on findings from other sessions.

Y Combinator
Building and demonstrating feature ideas — Allow agents to implement ideas and create screenshots or playable prototypes so weak ideas can be discarded before consuming human attention.

Y Combinator
Writing concise, proactive responses — Make the agent feel less like a terminal and more like a friend while hiding model, context, and session complexity.

Y Combinator
Building a WhatsApp relay — Let Peter send messages between his Mac and WhatsApp to monitor and interact with coding agents remotely.

Y Combinator
Organizational monitoring — Agents can analyze meeting transcripts and data to identify broken processes, conflicts, dependencies, and ground truth across an organization.

a16z
Research, retrieval, and memory — Agentic systems can use retrieval and graph generation to determine which information and tokens to keep in context for a task.

a16z
Coding and software creation — Vibe coding and coding agents can multiply an individual's output and reduce the effort required to ship and test software.

a16z
Unattended coding iteration — Have worker and critic agents repeatedly implement and assess bounded coding goals.

ManuAGI - AutoGPT Tutorials
Browser automation — Operate websites through managed browser sessions, identities, credentials, and agent-driven workflows.

ManuAGI - AutoGPT Tutorials
Coding-agent execution and permission handling — Run coding tasks, request approvals, and continue work across local and remote interfaces.

ManuAGI - AutoGPT Tutorials
Knowledge graph construction and enterprise question answering — Connect scattered data, trace answers to facts, and support governed agent actions.

ManuAGI - AutoGPT Tutorials
Edit and operate software repositories from a terminal — Provide coding-agent features including language-server operations, debugging, persistent cells, web browsing, and sub-agents

ManuAGI - AutoGPT Tutorials
Agent development and evaluation — Prototype, trace, debug, replay, and evaluate agent runs while monitoring model calls and tool executions.

ManuAGI - AutoGPT Tutorials
Agent-powered vulnerability analysis — Scan large repositories, investigate candidate issues, produce findings and recommendations, and revalidate against Git history.

ManuAGI - AutoGPT Tutorials
Browser automation — Let agents use isolated browser spaces to snapshot, click, fill, and navigate without disturbing the user's tabs.

ManuAGI - AutoGPT Tutorials
Collaborating with humans in shared rooms — Allow agents to open repositories, send patches, review code, run workflows, and edit canvases alongside people.

ManuAGI - AutoGPT Tutorials
Browser-game development — Three.js Game Skills routes agents across gameplay, graphics, UI, debugging, QA, and asset generation, with evidence-based checks before completion.

ManuAGI - AutoGPT Tutorials
Knowledge retention — self-learning-skills teaches coding agents to save reusable procedures and debugging lessons for later sessions.

ManuAGI - AutoGPT Tutorials
Verified iterative execution — loop.js runs an agent repeatedly until a separate read-only verification agent accepts the result.

ManuAGI - AutoGPT Tutorials
Remote agent operation — lfg provides a private control plane for starting, monitoring, and steering coding agents away from the machine hosting repositories and credentials.

ManuAGI - AutoGPT Tutorials
Persistent agent memory — Cognee stores agent knowledge in a self-hosted graph combining semantic embeddings and graph relationships.

ManuAGI - AutoGPT Tutorials
Web-interface control — Page Agent lets users control a web interface with plain language through an in-page JavaScript agent.

ManuAGI - AutoGPT Tutorials
Codebase understanding — codebase-memory-mcp lets AI coding agents query a persistent graph of code structure and relationships.

ManuAGI - AutoGPT Tutorials
Parallel coding-agent execution — Orca runs multiple CLI coding agents in isolated Git work trees for comparison and merging.

ManuAGI - AutoGPT Tutorials
Interface design and implementation — UI Skills gives AI coding agents task-specific guidance for user-interface work.

ManuAGI - AutoGPT Tutorials
Running terminal coding agents — Open Interpreter adapts agent scaffolding to low-cost models and supports MCP, skills, hooks, and permissions.

ManuAGI - AutoGPT Tutorials
Incident response — An AI agent maps infrastructure, groups related alerts, suppresses noise, traces blast radius, and diagnoses root causes.

ManuAGI - AutoGPT Tutorials
Persistent context and retrieval — Shared memory layers and knowledge graphs give agents access to current decisions, documents, browsing history, and cited passages.

ManuAGI - AutoGPT Tutorials
Building a personal context model — Persome converts permitted macOS activity into an inspectable, evidence-backed model that agents can access through MCP.

ManuAGI - AutoGPT Tutorials
Governed agentic development — Clodex combines models, code, terminal, browser, Git, and memory while controlling sensitive actions through permissions and a guardian.

ManuAGI - AutoGPT Tutorials
Executing coding agents in isolated environments — Clawk runs coding agents inside disposable network-restricted Linux VMs.

ManuAGI - AutoGPT Tutorials
Delegating coding work to subagents — PI Subagents creates separate foreground or background sessions for code review, codebase research, planning, and other roles.

ManuAGI - AutoGPT Tutorials
Natural-language financial research and backtesting — Vibe-Trading turns finance questions into analysis, strategy backtests, reports, and multi-agent investment, quant, and risk workflows.

ManuAGI - AutoGPT Tutorials
Managing and coordinating agent teams — LobeHub lets users build, schedule, group, and receive reports from AI agents.

ManuAGI - AutoGPT Tutorials
Agent browser automation — Moli provides a headless browser engine that builds DOM layout on demand for geometry or screenshots.

Github Awesome
Agent communication — Pi-peer lets coding-agent sessions on one machine exchange constrained plain-text messages.

Github Awesome
Design assistance — Airship lets an existing coding agent edit real project source from described visual changes.

Github Awesome
Coding-agent orchestration — Waku places multiple local coding agents behind one interface and maintains their sessions and project tasks.

Github Awesome
Coding assistance — Ante runs a terminal coding-agent loop offline using its embedded inference engine.

Github Awesome
Research software architectures — Make a coding agent consult arXiv papers and converge on a recommendation with documented prior failure modes.

Github Awesome
Evolve coding agents — Preserve, mutate, or replace agents' prompts, code, tools, and dependencies based on benchmark evidence.

Github Awesome
Run persistent coding workflows — Preserve sessions, memory, schedules, checkpoints, and reusable lessons while delegating work to isolated sub-agents.

Github Awesome
Modify sandboxed applications — Allow agents to work on private per-user copies while queuing simulated permission outcomes for bulk approval.

Github Awesome
Compress coding-agent context — Reduce upstream tool-output and history context while retaining expandable references.

Github Awesome
AI-agent workflow evaluation — RealReplicaBench tests whether agents can complete 107 long business workflows in reproducible service replicas.

Github Awesome
Personal assistant memory retrieval — Waku-agent demonstrates an agent harness, loop, memory, and evaluation structure while deciding whether each turn needs memory.

Github Awesome
Agent file-system execution — Cloudflare Computer gives an agent a durable SQLite-backed file system projected into containers, shells, or ECMAScript workers.

Github Awesome