MailsRAG
AI email automation for support and sales teams
One pattern we kept seeing while building MailsRAG is that support and sales teams do not usually fail because the model is weak. They fail because the workflow lets the model invent answers from the incoming email alone.
That creates replies that sound fine but drift away from pricing, policy, product facts, or security boundaries.
So we focused on a retrieval-first workflow:
- search the most relevant company knowledge first
- draft from that evidence
- keep review-first approval in place
- expand automation only where the results stay stable
We wrote up the reasoning here:
https://www.mailsrag.com/en/knowledge-based-email-replies/
Product page:
https://www.indiehackers.com/product/mailsrag
Also on dev.to:
And on Medium:
If you are working on AI email automation, I would love to hear whether your team trusts generic drafts or grounded drafts more in real inboxes.
Hi everyone, we are building MailsRAG for teams that want AI email automation without sending raw model output directly to customers.
What kept standing out to us is that most teams do not fail because the model is weak. They fail because the workflow is too loose. Once replies are generated from scratch without grounded knowledge or review, quality drifts quickly.
So our approach with MailsRAG is:
- retrieve shared knowledge first
- draft from that knowledge
- keep review-first approval
- support bilingual support and sales workflows
We wrote up the full thinking here:
https://www.mailsrag.com/en/how-to-automate-support-email-with-ai/
Product page:
https://www.indiehackers.com/product/mailsrag
Also published on dev.to:
https://dev.to/angsanhuang/how-to-automate-support-email-with-ai-without-losing-quality-control-536e
And on Medium:
https://medium.com/@angsanhuang/how-to-automate-support-email-with-ai-without-losing-quality-control
I would especially love feedback from anyone building for support teams, sales automation, or bilingual customer operations.
Comment