Configuring Monitoring and Traces

This guide will walk you through the process of setting up and configuring Tailwinds Monitoring for your AI workflows.

Step 1: Access Tailwinds Monitoring

1. Log in to your Tailwinds dashboard.

2. Click the "Monitoring" button in the navigation menu.

3. If you're not already authenticated, you may be prompted to log in again for security purposes.

Step 2: Create a Project

Projects in Tailwinds Monitoring allow you to organize and filter your workloads effectively.

1. Once logged in, you'll see the option to create a new Organization and Project.

   1. An Organization should be used for specific Teams, Departments, or Workloads

   2. You can have many Projects within an Organization, Projects should be used for different applications, or workload tiers (ex. Dev/Staging/Prod environment)

2. Click "Create Project" and give it a descriptive name (e.g., "General Use", "Customer Service Bot", etc.).

3. Your new project will appear in the project list.

Step 3: Configure Your Trace

Traces are at the heart of Tailwinds Monitoring, allowing you to track and analyze your AI workflows.

1. Within your project, click on "Configure Trace".

2. You'll be presented with configuration options and API keys.

API Keys

At the bottom of the configuration page, you'll find your API keys. These are crucial for integrating Tailwinds Monitoring with your AI applications.

• Public Key: Used for client-side tracing

• Secret Key: Used for server-side tracing (keep this secure!)

Integration Options

Tailwinds Monitoring offers several pre-made integration pathways for popular frameworks and languages. You'll find code snippets and instructions for:

• Python

• JavaScript/TypeScript

• LangChain

• OpenAI

• Anthropic

• And more...

Select the integration that best fits your tech stack and follow the provided instructions.

Step 4: Viewing Your Traces

Once you've set up tracing in your application, you can view your traces in the Tailwinds Monitoring dashboard.

1. Navigate to the "Traces" section in your project.

2. You'll see a list of recent traces, which you can click on for detailed information.

Here's an example of what a trace might look like:

Configuring Tailwinds Workflows to use your traces

Once your API keys are configured, please go to following page to configure and integrate your workflows.

Analytics

Additional Resources

Tailwinds Monitoring is built on top of LangFuse, an open-source LLM engineering platform. For more detailed information and advanced configuration options, you can refer to the LangFuse documentation.