Develop with AI. Verify the result.
Developer tasks with Claude, ChatGPT, and Gemini: from a clear request to coding agents and pipelines with human decisions.
Complete your first task Choose your tool Something's going wrong →
What are you struggling with?
- AI gives me generic answers Say what you actually want →
- It answered a different situation than mine It only knows what you paste →
- I'm on the fourth rewrite and it's not improving Make the next rewrite more useful →
- The answer looks right but I'm not sure it's true Check it before you act on it →
- I'm copying data around by hand, again You are doing work it could structure →
- I'm describing things it could just read Chat is not always the right tool →
The library behind it
Foundations
How these systems actually work, without the hand-waving.
10 pages
Everyday Claude
Chat, Projects, memory, files, plans and limits.
13 pages
Developing with ChatGPT
Reason from code and sources, maintain context, and return checked artifacts.
10 pages
Developing with Gemini
Engineering briefs, Gems, and research with selected sources.
8 pages
Claude Code
The terminal agent, for people who ship software.
19 pages
Working with Codex
From a clear task to a checked change, with reusable instructions and skills.
6 pages
Working with Gemini CLI
Repository context, planning, checked changes, and programmatic use.
3 pages
Cowork
An agent working across your own files and connectors.
14 pages
Anthropic API
API, agents, tools and MCP.
14 pages
OpenAI API
Bounded model calls, structured artifacts, and explicit access.
1 page
Gemini API
Direct model calls and validated handoffs for your own programs.
1 page
Trust and honest value
Whether to believe the output, and whether it is really helping.
7 pages
Security
Prompt injection, agent permissions, blast radius.
10 pages
Business and governance
Rollout, data rules, the AI Act, works councils.
12 pages
By context
Assembled playbooks for students, teams, regulated work.
12 pages
Developer workflows
Complete tasks, hand off checked artifacts, and automate stages across providers.
12 pages
Reference
Model matrix, glossary, quota, where to go instead of here.
5 pages
Recently verified
- Permissions: allow, gate, deny Shared methods Claude Code
- Skills, and when to write one Claude Code Claude Code
- Slash commands, and the prompt you keep retyping Claude Code Claude Code
- Subagents: delegating without losing the thread Claude Code Claude Code
- MCP servers in Codex: when they are worth it Codex Working with Codex
- MCP servers in Gemini CLI: when they are worth it Gemini CLI Working with Gemini CLI
- Official documentation, and when to go there instead Shared methods Reference
- Apps and plugins: what ChatGPT may read and what it may do ChatGPT Developing with ChatGPT