I’m building Orchestra to find the opportunities your AI agents can’t

I’ve spent the last two months building Orchestra around one simple idea: finding opportunities shouldn’t be what keeps you from getting the work.

Orchestra is an AI research agent that finds businesses, leads, jobs, decision-makers, and their direct contact details across the web. It can also search for local businesses, giving you a way to research opportunities that would otherwise take hours of manual searching, filtering, and cross-checking.

But finding the information is only the first half of what Orchestra does.

Orchestra has its own MCP server, which means it can be connected directly to the AI agents you already work with.

You can connect Orchestra to an agent such as ChatGPT, Claude, OpenClaw, Hermes, or any other compatible agent, and use them together as part of the same workflow.

For example, you could ask Orchestra to find marketing agencies in Berlin that may be relevant prospects. Orchestra researches them, finds the businesses and relevant decision-makers, collects their contact information, and returns the research.

You can then tell Orchestra to use a specific agent on your account and prepare an outreach workflow for those contacts.

Orchestra passes the research to that agent through MCP, along with the context it needs to continue the task. That agent can then handle the next step using the tools available to it, such as Gmail or another connected service.

The idea is that you don’t have to spend your time finding who to contact before your agents can start working.

• You give Orchestra the objective.

• Orchestra does the research.

• Your other agent takes it from there.

The end goal is simple: help you find opportunities for a deal, a client, a job, or whatever you’re looking for, without making you do all of the research yourself.

This is still under development, but after two months of building, the core system is finally coming together.

Here’s a real example of Orchestra researching:

I’m building this for people who already use AI agents and want them to do more than just wait for work to be handed to them.

What would you want Orchestra to find for your agents?

Would love to hear your feedbacks