Quickstart
Go from signup to your first AI-assisted debug session in minutes.
Get started with Sazabi and ask your first question about your production systems.
Prerequisites
Before you begin, make sure you have:
- A Sazabi account. Request access at sazabi.com
- An application that generates logs (any language or framework works)
Get started
Create your project
A project in Sazabi represents a single application or service you want to monitor.
- Log in to app.sazabi.com
- Click New project in the sidebar
- Enter a name for your project and click Create
# Install the CLI
npm install -g @sazabi/cli
# Log in to your account
sazabi auth loginAfter logging in, create a project from the dashboard at app.sazabi.com.
Connect your first data source
Sazabi needs access to your logs to answer questions about your systems. OpenTelemetry is the fastest way to get started.
All data sources send logs to Sazabi's backend by default. See Sending logs for more options including direct API ingestion and cloud provider integrations.
Add the OpenTelemetry SDK to your application and configure it to send logs to Sazabi:
# Set your Sazabi API key as an environment variable
export OTEL_EXPORTER_OTLP_ENDPOINT="https://otlp.us-west-2.intake.sazabi.com"
export OTEL_EXPORTER_OTLP_HEADERS="Authorization=Bearer YOUR_PUBLIC_API_KEY"You can find your API key in Settings > API Keys in the dashboard.
Ask your first question
Once logs are flowing, open your project in the Sazabi dashboard and start chatting. Chat understands your logs and can answer questions in natural language.
Try these example prompts:
What errors happened in the last hour?Show me the slowest API endpointsWhy is the checkout service returning 500s?Sazabi searches your logs, analyzes patterns, and responds with specific evidence from your data.
Explore the response
Sazabi shows its work as it investigates your question. Each response includes:
- Tool calls: Expandable sections showing the queries Sazabi ran and the data it found. Click any tool call to see the raw results.
- Evidence: Log entries, stack traces, and metrics that support the answer.
- Follow-up suggestions: Related questions you can ask to dig deeper.
You can continue the conversation naturally. Ask follow-up questions, request more detail, or pivot to a different area of your system.
Next steps
Connect more data sources
Add logs from additional services, cloud providers, or logging frameworks.
Set up Slack
Get alerts and ask questions directly in Slack channels.
Invite your team
Add teammates so everyone can investigate issues together.
Learn about alerts
Configure proactive monitoring to catch issues before users report them.