WaddleAI

Get started with WaddleAI

Kubernetes via Helm is the only supported deployment path. This deploys the full platform: proxy, management API, WebUI, PostgreSQL, and Valkey.

  1. 1. Clone the repo

    git clone https://github.com/penguintechinc/waddleai.git
cd waddleai

  2. 2. Create the namespace and install with Helm

    kubectl create namespace waddleai
helm install waddleai k8s/helm/waddleai \
  --namespace waddleai \
  --values k8s/helm/waddleai/values-beta.yaml \
  --set secrets.postgresPassword="$(openssl rand -hex 16)" \
  --set secrets.jwtSecret="$(openssl rand -hex 32)" \
  --set management.env.ADMIN_INITIAL_PASSWORD="$(openssl rand -hex 16)"

  3. 3. Check the rollout

    kubectl get pods -n waddleai
kubectl port-forward -n waddleai svc/waddleai-management 8001:8001 &
curl http://localhost:8001/healthz

  4. 4. Generate a virtual API key and connect

    Log in as admin to get a JWT, generate a virtual API key (wa-...), then point the OpenAI SDK, Anthropic SDK, or the Assistant at the proxy on port 8080. See the full Quick Start for a documented chart-gap workaround.

More resources