Weights & Biases — Weave

GigaFlow reads Weave calls via the W&B trace-server API. A template transform (wb_weave.yml) ships, but you'll likely tailor its filter rules to your op names (Weave has no structural span-type field).

What you'll need

• A WANDB_API_KEY (wandb.ai → Settings → API keys).
• Your Weave project id as entity/project (e.g. my-team/my-agent).
• Base URL https://trace.wandb.ai (default; override for self-hosted).

Connect

Run the setup wizard. The first time, it signs you in with your waitlist email (same as gigaflow login), then walks you through the connection above:

gigaflow setup

Pick W&B Weave when prompted and enter your entity/project and WANDB_API_KEY. The wizard creates a GigaFlow project, applies a transform, registers the datasource, and runs the first sync — you never set an API key or backend URL by hand.

Transform

Weave calls normalize to nested dicts with attributes.*, inputs, output, op_name, display_name. The bundled wb_weave.yml is a template — verify the classification preview the wizard shows, and if your op names don't match, copy the template, adjust the filter rules, and give the wizard the path to your edited file when it asks for a transform.

After the first sync

gigaflow sync                                    # re-pull new traces anytime
gigaflow compute "SELECT trace_id FROM trace_metrics WHERE run_id IS NULL"
gigaflow inspect <trace_id>