Integrations overview

Bidderops connects to the outside world in a few well-defined ways. This page is a map of those connections and, importantly, of where your data goes in each case.

The connections

Integration	Direction	What crosses the boundary
AI providers	Outbound	Opportunity facts, RFP text, pipeline digests, and your custom instructions are sent to the model you choose — which can be a cloud provider or a model running on your own hardware.
Local LLM	Outbound (to you)	The same AI features, but pointed at an endpoint you control, so the data never leaves your infrastructure.
Tender sources	Inbound	Public procurement notices are fetched from TED, the World Bank, UNDP, the IDB, and custom RSS feeds.
MCP server	Bidirectional	An AI assistant (such as Claude or Cursor) can securely read and act on your pipeline through authenticated tools.
Sign-in	Inbound	Authentication, including optional Google single sign-on.

What stays inside your organization

Everything you create in Bidderops — opportunities, documents, notes, outcomes, experience records — is stored against your organization and is invisible to other organizations. The only data that leaves Bidderops is what you deliberately send to an AI provider when you run an AI feature, and even that can be kept on your own infrastructure by running a local model.

Choosing your AI posture

You configure AI per feature in Settings → AI, and you can mix providers:

Use a cloud model with web search for bid intelligence, where current facts matter, and
Use a local model for privacy-sensitive features like Go/No-Go, where the prompt contains internal numbers.

The next pages explain each integration in detail.