As it is said here.
Clients can connect to the Private endpoint from the same VNet, peered VNet in same region, or via VNet-to-VNet connection across regions.
Additionally, clients can connect from on-premises using ExpressRoute, private peering, or VPN tunneling. Below is a simplified diagram showing the common use cases.
You can create a self-hosted agent on the azure SQL server VM, or on an Azure VM which is connected to the same VNet. Check here for detailed steps to create self-hosted agent.
You can also create a self-hosted agent on your local machines. And using ExpressRoute, private peering, or VPN tunneling to connect to the Private Endpoint. For more information, check the document Connecting from an on-premises environment over VPN
Then you can run your pipeline on the self-hosted agent by setting the pool to your private agent pool which hosts the self-hosted agent. Your pipeline should be able to connect to the SQL server, since the agent machine is configured to connect to the Private Endpoint.