I am exploring how generative AI is changing the way we build software, using tools like Claude Code, GitHub Copilot and the OpenAI Codex CLI to design, build and review real systems, and running open models locally with Ollama, while keeping a human firmly in the loop as the quality control.
Setup guides
Step-by-step guides to getting each AI coding tool running:
How to Set Up Claude Code
Hands-on with Anthropic’s agentic coding tool for the terminal: install, sign in, and your first session. Read the guide →
How to Set Up GitHub Copilot
Get the AI pair-programmer running in VS Code and JetBrains, with inline suggestions and chat. Read the guide →
How to Set Up the OpenAI Codex CLI
Set up OpenAI’s terminal coding agent: install, sign in, and stay in control with approval modes. Read the guide →
Running models yourself
How to run large language models on your own hardware and in the cloud:
How to Run AI Models Locally and in the Cloud with Ollama
Run open models on your own machine with Ollama, and offload the big ones to the cloud. Read the guide →
More experiments, notes and lessons coming soon.