From KachiluScout to Kachilu: controlled AI outbound workflows

Hey Product Hunt 👋　I’m the maker of Kachilu.

Kachilu started from a much narrower product called KachiluScout — a LinkedIn GTM tool for recruiting. The original idea was to help teams find the right candidates, send connection requests, and follow up with DMs.

But as we worked with users, we realized that was not enough.

For high-quality outbound, the first message is only one part of the workflow. In many cases, people get much better acceptance and reply rates when they engage with the target first — for example by understanding their profile, interacting with their posts, leaving thoughtful replies, and then reaching out with more context.

We also heard two clear signals from users:

First, they did not want to be limited to LinkedIn. Many teams wanted to run workflows across other social and community channels too.

Second, some users were already using the product beyond recruiting — for sales, marketing, founder-led GTM, and agency workflows.

That is why we rebuilt Kachilu as a broader AI browser agent for controlled outbound workflows.

Kachilu helps teams plan and execute browser-based tasks across social, community, and web channels — including lead research, post engagement, outreach preparation, form outreach, follow-ups, and execution tracking.

Because outbound automation can easily become risky if it is not designed carefully, we put a lot of focus into control and safety: task-level instructions, schedules, execution limits, pause/resume, human handoff, and full execution history.

Our goal is not blind automation.

Our goal is to help teams execute more consistently while staying in control.

Would love your feedback:

What parts of outbound workflows would you trust an AI agent to execute — and where should humans always stay in the loop?