VectorSurfer
Getting Started
Sign up, connect your Weaviate instance, and start monitoring.
Prerequisites
Before using VectorSurfer, you need:
- VectorWave installed and configured in your Python project (Installation)
- A running Weaviate instance with VectorWave data
- A modern web browser
Step 1: Sign Up
Visit the VectorSurfer dashboard and create an account:
- Click Sign Up
- Enter your email and password
- Optionally add a display name
- Click Sign Up to create your account
Already have an account? Click Log In and enter your email and password.
Step 2: Create a Weaviate Connection
After signing in, you'll be directed to the Projects page. This is where you connect VectorSurfer to your Weaviate instance.
- Click Create Connection
- Choose a connection type:
| Type | Description |
|---|---|
| Self Hosted | Your own Weaviate instance (Docker, Kubernetes, etc.) |
| WCS Cloud | Managed instance on Weaviate Cloud |
- Fill in the connection details:
| Field | Default | Description |
|---|---|---|
| Name | Default | A label for this connection (e.g., "Production") |
| Host | localhost | Weaviate server hostname or IP |
| HTTP Port | 8080 | Weaviate REST API port |
| gRPC Port | 50051 | Weaviate gRPC port |
| API Key | — | Authentication key (required for WCS Cloud, optional for self-hosted) |
- Click Test Connection to verify connectivity
- If the test passes, click Create to save the connection
- Click Activate on the connection card to make it active
First time? A quick-start guide appears at the bottom of the Projects page. Click Start Guide for a step-by-step walkthrough covering Weaviate setup and VectorWave installation.
Step 3: Set Up VectorWave
If you haven't already, install VectorWave and start logging function executions:
pip install vectorwave
from vectorwave import vectorize, initialize_database
initialize_database()
@vectorize(auto=True, capture_return_value=True)
def your_function(prompt: str):
# Your AI function logic
return result
Once functions are decorated with @vectorize, every execution is automatically stored in Weaviate. VectorSurfer reads from the same Weaviate instance and displays your data immediately.
See the VectorWave Quick Start for detailed setup instructions.
Step 4: Configure AI Features (Optional)
AI-powered features like Ask AI, Healer diagnosis, and semantic search require an OpenAI API key.
Global API Key
- Go to Settings (gear icon in the sidebar)
- Open the AI Settings tab
- Enter your OpenAI API key (
sk-...) - Click Save
This key is used for all AI features across all connections.
Per-Connection API Key
You can also set a separate OpenAI key for a specific connection:
- Go to Settings → Connections tab
- Click the key icon next to a connection
- Enter the API key and save
Connections with an API key configured show an AI Ready badge on their card.
AI features are optional — the dashboard works for monitoring, tracing, and execution browsing without an OpenAI key.
Step 5: Explore the Dashboard
Once your connection is active and VectorWave has recorded executions, the dashboard displays your data:
- KPI Cards — Total executions, success rate, error rate, cache hit rate, average duration
- Execution Timeline — Function activity over time
- Recent Errors — Latest errors at a glance
- System Status — Weaviate connection health
The dashboard is fully customizable — click the Edit button to add, remove, resize, and reorder widgets. See Usage Guide for details on all available widgets.
Next Steps
- Usage Guide — Learn how to use each dashboard feature
- VectorSurfer Overview — Full feature list
- VectorSurferSTL — Lightweight Python-only alternative