Choose the job, then examine the tool’s capabilities, data access, and limits.
Inside the platforms
Choose a platform to see its components, guided examples, and complete workflows. Consumer assistants, coding environments, developer APIs, and automation builders have different boundaries.
Open the Platform Deep Dives index
Conversation & research
Grok
Understand Grok’s chat, search, files, and creative tools—and the different boundaries of standalone access, X, Bot, and the API.
Meta AI
Start with Meta’s everyday assistant, then understand how it relates to Muse, Llama, connected apps, and AI glasses. Two small walkthroughs show where to begin.
ChatGPT
An assistant and agent workspace for explanations, research, files, and connected work.
Claude
A conversational and agentic workspace for sourced reasoning, documents, reusable procedures, and deliverables.
Gemini
A Google AI workspace for research, connected sources, editable drafts, notebooks, media, and bounded recurring work.
Microsoft Copilot
A family of personal and work AI experiences whose identity, grounding, editing, and agent boundaries matter.
Perplexity
A sourced-answer and research platform that also organizes projects and orchestrates deliverable-producing agents.
Coding & development
Codex
A coding agent for repository inspection, scoped changes, command execution, and reviewable development work.
Claude Code
A development agent whose project context, extensions, and enforceable tool permissions need separate understanding.
GitHub Copilot
An editor and repository assistance family spanning suggestions, interactive agents, cloud work, and code review.
Cursor
A coding environment that connects agent context, local tools, rules, extensions, and prepared cloud development.
Workflow & automation
n8n
A workflow builder that connects deterministic data flow with model calls, agents, knowledge, and human review.
Make
A visual automation platform for fixed scenarios and beta agents with explicit data, tool, and usage boundaries.
Zapier
A connected workflow platform whose AI steps, tool agents, approvals, and stored records should remain distinguishable.
Power Automate
A Microsoft workflow system spanning cloud actions, desktop automation, AI prompts, approvals, and governed environments.
More tool starting points
Gemini Notebook
Work with a source collection and turn it into explanations, briefings, and learning material you can inspect.
Perplexity
Explore web-backed answers and source discovery while checking whether the cited material supports the conclusion.
GitHub Copilot
Use coding assistance and agent features inside a development workflow with repository context and human review.
Cursor
Explore an AI coding environment with project context, agent features, and configurable instructions or connections.
Codex
Explore OpenAI’s coding agent for repository tasks, proposed changes, and reviewable development work.
Adobe Firefly
Generate and edit visual material with an explicit creative brief and a review of rights, fidelity, and audience interpretation.
Midjourney
Explore visual generation through prompts, references, and iteration, with a clear decision about how the media will be used.
Runway
Explore generative video and media workflows while checking motion, continuity, rights, and the meaning of the scene.
ElevenLabs
Explore speech generation, transcription, and voice applications with consent and accurate representation at the center.
Descript
Explore transcript-centered audio and video editing, then review both the words and the original media.
n8n
Build visible workflows that connect apps, AI steps, and tools, with explicit error paths and operating responsibility.
Make
Connect apps with visual scenarios, distinguishing fixed automations, AI steps, and model-directed agent paths.
Zapier
Connect a trigger to app actions, then add AI only where interpretation or generation improves the process.
Microsoft Power Automate
Explore automated processes across Microsoft and other services with defined ownership and approval points.
OpenAI Agents SDK
Build model-and-tool systems in code with explicit control over runs, handoffs, state, and review.
LangGraph
Coordinate stateful workflows and agents with a clear distinction between fixed steps and model-driven choices.
OpenAI API
Use models and tools inside your own application while owning the interface, credentials, data handling, and operating limits.
Claude API
Explore Anthropic’s model platform as a component in an application, with task-specific testing and defined tool access.
Gemini API
Explore Google’s model platform for text and multimodal application tasks with clear inputs and review criteria.
Hugging Face Hub
Explore models, datasets, and application examples while checking model cards, licenses, provenance, and task fit.
Ollama
Explore running models on your computer or through supported services, checking which mode the task actually uses.
LM Studio
Explore a desktop interface for local models, then inspect the services and connections you enable around them.
scikit-learn
Build and evaluate models for classification, regression, clustering, and other structured-data tasks.
PyTorch
Explore a programming framework for training and using neural networks with datasets, optimization, and evaluation.
ChatGPT
A starting point for conversational learning, drafting, and exploring tasks.
Microsoft Copilot
Explore assistance within Microsoft’s productivity ecosystem.
Choose a tool by the task
A decision framework is more useful than a universal “best AI” ranking.