Load Kubernetes Gateway API Inference Extension data to DuckDB

The Kubernetes Gateway API Inference Extension provides declarative Kubernetes-native APIs to optimize and standardize routing for self-hosted generative AI models on Kubernetes-based infrastructure. Everything needed to build a working Kubernetes Gateway API Inference Extension → DuckDB pipeline is on this page: the API's base URL, authentication, endpoints, pagination and incremental field — plus a prompt that hands the whole job to your coding agent.

---

Build your Kubernetes Gateway API Inference Extension to DuckDB pipeline

Paste this prompt into Claude, Codex, or Cursor. The agent does the rest.

PromptCopy prompt
Run uvx dlthub-init@latest to build a pipeline from Kubernetes Gateway API Inference Extension to DuckDB and run it on dltHub

That scaffolds a dltHub workspace and installs the dltHub AI harness — the project rules, the secrets-management skill, and the dlt MCP server your agent needs to work safely. From there it reads the Kubernetes Gateway API Inference Extension API, proposes the endpoints to load, then writes, runs and validates the pipeline while you review rather than type. Credentials are inspected through MCP tools, so your agent never reads secrets.toml itself. How the LLM-native workflow works →

Prefer to write it yourself? Every fact the agent uses is below.

---

Kubernetes Gateway API Inference Extension API at a glance

These values come from the Kubernetes Gateway API Inference Extension API reference — the authoritative source if anything here looks out of date.

---

How do I authenticate with the Kubernetes Gateway API Inference Extension API?

Authentication is handled at the gateway layer (e.g., Envoy Gateway, kgateway, GKE Gateway) using standard Gateway API patterns such as JWT, OpenID Connect, or Mutual TLS via BackendTLSPolicy, rather than by the Inference Extension itself.

1. Get your credentials

The Kubernetes Gateway API Inference Extension is a Kubernetes-native open-source project that runs inside your cluster. It does not have a central cloud-hosted "dashboard" or a single REST API that requires global API keys in the way a SaaS product would. Instead, authentication and API management are handled by the underlying Gateway implementation you use (e.g., GKE Inference Gateway, Envoy Gateway, or NGINX Gateway Fabric). To obtain credentials: 1. If using GKE Inference Gateway, configure an Apigee backend for API management, where you create API keys via the Apigee UI. 2. For self-hosted deployments, API security is typically managed via standard Kubernetes mechanisms (e.g., ServiceAccounts, RBAC) or by integrating the gateway with an external AI gateway provider like LiteLLM or Apigee. Check your specific gateway implementation documentation for its authentication configuration.

2. Add them to .dlt/secrets.toml

Copy
[sources.kubernetes_gateway_api_inference_extension_source]
api_key = "your_bearer_or_api_token_here"

dlt reads this file automatically at runtime. With the harness, the setup-secrets skill prompts you for the values and never handles the raw credential in chat. For production, see setting up credentials with dlt.

---

What Kubernetes Gateway API Inference Extension data can I load into DuckDB?

These are the Kubernetes Gateway API Inference Extension endpoints dlt can load into DuckDB:

---

How do I load only new Kubernetes Gateway API Inference Extension records?

The Kubernetes Gateway API Inference Extension API reference does not document a timestamp or sequence field for these endpoints, so there is nothing to advertise here as verified. Pick a field from the endpoints table above that increases with every write, then set it as the cursor_path.

Copy
{"name": "inference_pools", "endpoint": {
    "path": "apis/inference.networking.k8s.io/v1/inferencepools",
    # Replace with a field that increases on every write.
    "incremental": {"cursor_path": "REPLACE_ME", "initial_value": "2024-01-01T00:00:00Z"},
}}

On the first run dlt loads everything from initial_value; on every run after that it requests only what changed and appends with write_disposition="merge" if you set a primary key. See incremental loading.

---

What does the generated Kubernetes Gateway API Inference Extension pipeline look like?

A standard dlt REST API pipeline — the same code you would write by hand, loading /v1/completions and /v1/chat/completions from the Kubernetes Gateway API Inference Extension API into DuckDB:

Copy
import dlt
from dlt.sources.rest_api import RESTAPIConfig, rest_api_resources

@dlt.source
def kubernetes_gateway_api_inference_extension_source(not_applicable=dlt.secrets.value):
    config: RESTAPIConfig = {
        "client": {
            "base_url": "Not applicable (The project provides Kubernetes CRDs/APIs for configuration, not a central REST API service)",
            "auth": {"type": "bearer", "token": not_applicable},
        },
        "resources": [
            {"name": "inference_pools", "endpoint": {"path": "apis/inference.networking.k8s.io/v1/inferencepools"}},
            {"name": "inference_pool_imports", "endpoint": {"path": "apis/inference.networking.k8s.io/v1/inferencepoolimports"}}
        ],
    }
    yield from rest_api_resources(config)


def load_kubernetes_gateway_api_inference_extension_to_duckdb() -> None:
    pipeline = dlt.pipeline(
        pipeline_name="kubernetes_gateway_api_inference_extension_pipeline",
        destination="duckdb",
        dataset_name="kubernetes_gateway_api_inference_extension_data",
    )
    load_info = pipeline.run(kubernetes_gateway_api_inference_extension_source())
    print(load_info)


if __name__ == "__main__":
    load_kubernetes_gateway_api_inference_extension_to_duckdb()

Run it with python kubernetes_gateway_api_inference_extension_pipeline.py. The agent iterates on this until it loads cleanly — you review and approve, rather than write it from scratch.

---

How do I query Kubernetes Gateway API Inference Extension data in DuckDB?

dlt creates one table per resource. Query the loaded data with Python or SQL — or ask your agent to, through the MCP server's execute_sql_query tool.

Python (pandas DataFrame):

Copy
import dlt

data = dlt.pipeline("kubernetes_gateway_api_inference_extension_pipeline").dataset()
df = data.inference_pools.df()
print(df.head())

SQL:

Copy
SELECT * FROM kubernetes_gateway_api_inference_extension_data.inference_pools LIMIT 10;

See querying your data with dataset and exploring it in marimo notebooks.

---

How do I deploy the Kubernetes Gateway API Inference Extension to DuckDB pipeline in production?

The pipeline runs locally, which is ideal for prototyping and one-off analysis. When you need it on a schedule, monitored on every load, and shared with your team, deploy the same dlt code on the dltHub platform — no infrastructure to maintain. The prompt above already ends with "run it on dltHub", so your agent can take it there directly.

• Deploy & schedule — run the pipeline as a managed job with automatic retries.
• Monitor — observable job queues, alerting, and load metrics for every run.
• Transform — promote raw Kubernetes Gateway API Inference Extension loads into governed, documented models.
• Visualize & share — explore data in notebooks and publish live dashboards instead of static screenshots.

Book a demo →

---

What other destinations can I load Kubernetes Gateway API Inference Extension data to?

dlt loads into any of these — only the destination argument changes:

Set dlt.pipeline(destination="snowflake") and add credentials in .dlt/secrets.toml. On the dltHub platform the same pipeline runs against a managed Iceberg lakehouse. See the full destinations list.

---

Next steps

• Ready for production? Book a demo of the dltHub platform →
• dlt LLM-native workflow docs
• REST API source reference
• Join the dlt community on Slack