AI
The pipeline decides
Forty PRs in six days. Five thousand lines of net change. Zero direct pushes to main. The model writes the code. The pipeline decides if it ships. The morning gate decides if it stays shipped.
Field CTO @ Gravitee
AI
Forty PRs in six days. Five thousand lines of net change. Zero direct pushes to main. The model writes the code. The pipeline decides if it ships. The morning gate decides if it stays shipped.
Agents
An agent reads my calendar every Monday, classifies every event, and combines it with GitHub, GA4, and task data into one dashboard. Eight weeks in: my mental model of my own time was wrong by thirty percent. You could build the same thing by Friday.
AI
You do not need a team, a budget, or a cloud contract to run AI in production. You need an afternoon, some thinking space and a GitHub Pro subscription. A Starlette 1.0 bump removed `add_event_handler`. It landed in a Dependabot PR at 03:00 on a Tuesday.
a podcast generated by AI to explain Gravitees 4.11 release
Existing tools, used differently
Technical
Rate limits, guard rails, plan-gated data access, semantic caching: all enforced at the gateway without touching a single line of agent code. A full technical walkthrough of the RNLI AI Agent demo.
AI
Policy documents do not stop a rogue agent. Neither does a terms of service. What actually enforces governance is the infrastructure layer — and this is what building it looks like in a mission-first organisation.
Agents
The Interconnect has an MCP server. Any agent with the right configuration can read, search and cite every article directly — no scraping, no guessing. Here is how it works and why agent-first publishing matters.