Developer overview

A familiar API with a stricter runtime behind it.

Connect with an OpenAI-compatible client, then layer in customer scope, governed knowledge, agents, metering, and audit as the product requires.

Request technical access Platform overview

First request

Change the base URL. Keep the client.

The credential represents the user and organization calling through your application. Model access and balance are checked before provider spend.

PythonOpenAI SDK

from openai import OpenAI

client = OpenAI(
    base_url="https://engine.example.com/v1",
    api_key="<tenant-token>",
)

response = client.chat.completions.create(
    model="auto",
    messages=[{"role": "user", "content": "Summarize the policy."}],
)

Technical paths

Go deeper by control boundary.

Public API reference is provided with technical access. These guides explain the product and operating model behind the endpoints.

Start

Gateway

Point an OpenAI-compatible client at Deeplinq and send the caller’s tenant credential.

Control

Governance

Understand roles, model grants, dataset grants, guardrails, and approval boundaries.

Build

Knowledge

Ingest governed data and return cited answers from only the datasets the caller may use.

Automate

Agents

Configure an assistant, start a durable run, and handle approval-required tool calls.

Operate

Audit trail

Verify chain integrity and export bounded audit evidence for independent review.

Next step

Bring your current client and deployment constraints.

We will map the shortest integration path and identify the controls that need configuration before the first production request.

Request technical access