Why I built Inbxer
Every B2B helpdesk I tried was built for someone else's problem. Here's the gap that made me start writing Inbxer, and why it had to be open source from day one.
AbdelmjidBuilding InbxerCopy linkThe pattern is always the same. A customer writes in about something that looks like a bug. Support asks for context the product already knows. Someone copies the thread into an issue tracker. The fix ships. Nobody tells the customer. I've watched that loop happen on enough teams, and used enough tools that made it worse instead of better, that I started writing my own.
What I actually wanted
Not another ticket queue — a support platform that understands who's writing. B2B support isn't anonymous: you have named accounts, each with a tier, history and relationship your team needs to understand. I wanted email, Slack and in-app chat to become one continuous customer history, not three disconnected tools with three separate login screens.
And I wanted the answer to reach engineering without a copy-paste. Half of what support needs to say about a bug lives in a codebase, not in a macro library.
What was actually on the market
Every helpdesk I evaluated was optimized for a different problem: high ticket volume, low context per contact, fast to close. That's the right shape for consumer support. It's the wrong shape for a team with a few hundred accounts where the answer to "why is this broken" is worth taking ten extra minutes to get right.
So I started writing it
Inbxer is a complete support platform built with Laravel 13 and React 19 — conversations, customer accounts, SLAs, reporting, a rule-based workflow engine, and AI agents that draft, summarize and organize work without taking the human out of the loop. I picked MIT over a more restrictive open-source license on purpose: I want people to run it, fork it, and embed it without a legal review, and I want any hosted version of it to win customers on being good, not on a license technicality.
Where it's headed
It's pre-launch. The conversation workspace, AI agents, automation, reporting, and the email, Slack and chat-widget channels all work today, and I run this site's own support on the same build. Next is the public v1 release, with a hosted Cloud version to follow for teams who'd rather not run the infrastructure themselves.