Horaizon Live: Building with Claude Code
A working session on where Anthropic's Claude Code fits into real engineering work — and where it still needs a hand on the wheel.
Luke Gregory works through how the Horaizon team uses Claude Code — Anthropic's terminal-native coding agent — inside client engagements. The setup, the patterns we now reach for by default, and the guardrails that keep an agentic loop grounded in the codebase it is supposed to be helping.
This isn't a demo. It's the same conversation we have with clients when they ask whether Claude Code will actually save their team time — the honest version, with the caveats.
[Vigtigste pointer]
- Where Claude Code compresses delivery time — and where it does not.
- Guardrails that keep an agentic coding loop grounded in the codebase.
- The review discipline that makes AI-generated code safe to merge.
- What to change in your team's review process before adopting it.
Vil du have den samme samtale om jeres arbejde?
Fortæl os, hvor AI ville tjene sin plads i din forretning. Vi vender tilbage med en scoping-plan, en realistisk tidslinje og et klart billede af omkostningerne.