We built Faultline at Chatwoot to help our team debug infrastructure incidents faster. It's an AI agent you can ask questions in plain English — it autonomously queries your monitoring tools, cloud infrastructure, and codebase, then comes back with a root cause analysis.
It connects to New Relic, Sentry, AWS, GitHub, and PagerDuty using MCP (Model Context Protocol) servers and native integrations. You ask something like "Why is the API latency high?" and the agent loops through your tools — pulling APM metrics, checking error rates, scanning recent deploys — until it has enough data to give you an answer.
Stack: Rails 8 + Sidekiq backend, Python/FastAPI agent service using OpenAI, Vue 3 frontend, PostgreSQL, Redis. All integrations are configured through the UI, credentials encrypted at rest.
It doesn't replace an SRE — but it does the first pass of pulling data from five different dashboards and correlating timelines, which usually takes 10-15 minutes manually.
Thanks for all the support you folks gave us in the past[1]. At Chatwoot[2], we are developing an open-source[3] customer support platform as an alternative to Intercom and Zendesk. I'm glad to announce our latest AI update. We have been working with the latest LLMs to see if we can assist support staff in reducing their workload.
We have observed large-scale Chatwoot deployments with over 10 million messages. From our discussions with the teams handling these deployments, we noticed two types of queries. The first type consists of simple questions, sometimes with minor tweaks. The second type requires human intervention and typically involves more complex support queries. Many tools first employ an FAQ bot, now an AI bot, before transferring to a human. This process can be frustrating for both the user and the support agent. We have found that customers, especially when they are already frustrated, dislike having to interact with a bot before reaching a human to resolve their issue.
The issue we face is the lack of distinction between the two types of queries. Not all queries are inherently support queries. Sometimes, the customer may not want to search through the documentation, or the documentation may not be well-organized enough to find the answer. Having a chat common chat interface makes it harder for everyone.
We believe that a clear UX distinction can make a significant difference. Here is how we plan to split these conversations.
a/ *Universal search interface*: This is a Chat GPT-like chat that you can integrate into your application. Users can click on the search icon or use CMD+K to access it. They can find answers using your help desk articles and carry out tasks through natural language queries like "Where is my order?" or "When is the next billing date?". We have found this can improve the quality of support requests received.
b/ *Live Chat SDK*: To address complex issues, your customers can communicate directly with you with the live chat SDK.
This approach will clarify expectations for the end customer. Use the search function for straightforward queries and engage with an agent for more complicated ones.
Along with these, we also introduced some updates to the agent interface. These updates will help agents prioritize conversations by auto-tagging them and routing them to the appropriate teams.
We are launching it as a closed beta at the moment to work closely with teams with scale and unique problems. Also, we have not yet figured out how we can add support for the open-source LLMs in different environments, as our installations are one-click at the moment. Adding those would make it complex for everyone.
Most people see this as some sort of tax to provide better support, which should not be the case. We need better tools at reasonable and transparent pricing. Hope we succeed in that mission.
Chatwoot is currently suited for support use cases. We are working on the customer success workflows (onboarding, retention etc) now. Hope it gets there soon.
The FB/Instagram channels are a pain to manage (even for us) due to the app review cycle. This is a common problem in the community. Recently we have seen many rejections in Instagram API reviews without a specific reason. Some of the alternatives the Meta developer review team suggests are also not working. We are trying to make this flow better using a hosted app or something. Open to ideas here.
We are aware of the issues with the mobile app. It is getting a revamp now, hope most of the issues are resolved by it.
We are following open core model. The idea is to build a sustainable way to provide more value in opensource. Bulk of the features are under MIT license.
@pranav_rajs I couldn't see some enterprise-grade features like audit logs or privacy vault. Would love to help for free. Open source & plug-n-play: www.boxyhq.com
We have 2 versions of software. Enterprise edition and the community edition. The repository has a dual license. The /enterprise folder is licensed under a proprietary license whereas everything else is under MIT license
We launched Chatwoot a couple of months back in HN and was well received by the community.
### What is Chatwoot?
Chatwoot is an open-source, self-hosted customer engagement suite, built as an alternative to Intercom and Zendesk. With Chatwoot, you can view and manage your customer data, communicate with them irrespective of which medium they use, and re-engage them based on their profile.
### What's new?
Today we are excited to make 3 major announcements:
1. 2.0 version
2.0 version ships with an updated dashboard that is completely redesigned for a better user experience. This update ships with many new features, including more communication channels, productivity enhancements, better conversations visibility, re-engagement options, etc.
2. Enterprise edition
We will include the features such as customisable dashboards, SLA Management, Agent scheduling software, IP blocklisting etc., suited for large organizations in the Enterprise Edition software. In contrast, Small and medium businesses would continue to use the Community edition. We are making the license update in the repository to include a dual licensing option.
3. Contributor Hall of Fame
The community has been core to our success so far. It is rewarding to see people from different countries and cultures come together for a common goal. Our product supports more than 30 languages right now, and all of it was contributed by the community. We want to honor and extend our gratitude to the people who have put the effort in helping us to make this software better every day. We would be publishing the contributor hall of fame.
We built Faultline at Chatwoot to help our team debug infrastructure incidents faster. It's an AI agent you can ask questions in plain English — it autonomously queries your monitoring tools, cloud infrastructure, and codebase, then comes back with a root cause analysis.
It connects to New Relic, Sentry, AWS, GitHub, and PagerDuty using MCP (Model Context Protocol) servers and native integrations. You ask something like "Why is the API latency high?" and the agent loops through your tools — pulling APM metrics, checking error rates, scanning recent deploys — until it has enough data to give you an answer.
Stack: Rails 8 + Sidekiq backend, Python/FastAPI agent service using OpenAI, Vue 3 frontend, PostgreSQL, Redis. All integrations are configured through the UI, credentials encrypted at rest.
It doesn't replace an SRE — but it does the first pass of pulling data from five different dashboards and correlating timelines, which usually takes 10-15 minutes manually.
Demo: https://youtu.be/S1-pW_wD2uA
GitHub: https://github.com/chatwoot/faultline
Blog: https://faultline.chatwoot.com/blog/introducing-faultline
Would love your feedback.