From the course: Claude Code: Designing Multi-Model AI Systems

Unlock this course with a free trial

Join today to access over 25,500 courses taught by industry experts.

Run the full stack: Debug view and metrics dashboard

Run the full stack: Debug view and metrics dashboard

From the course: Claude Code: Designing Multi-Model AI Systems

Start my 1-month free trial Buy for my team

Run the full stack: Debug view and metrics dashboard

“

We have built every piece of the system separately. Now let's stress test it. I'm going to build an integration test that throws realistic traffic at the full pipeline, not 5 messages, but 20 plus, and shows a live dashboard as it processes them. Here is the prompt. Let's go to the prompts folder and find 5.1.fullstack.md. Here is the prompt. We will send 20-plus messages with realistic distribution that will include eight product questions, five greetings, three off-topic, two complex, two adversarial, and two personal information. We will process each through the full pipeline, which means we will import from helpdesk bot or reimplement, guardrails, classify, route, and generate. Then, we will show a live dashboard that updates after each message. It will have the progress percentage. It will have the route counts updating, for example, SONET 4, CAN 5, and so on. It will have the running costs, very important. And it will have health, all green, or we may show any amber-red in…

Contents