RootCauseAI, our own product
Incident investigation that stays in your network
Describe the incident. RootCauseAI searches your code, logs, tickets and databases, then shows the likely cause and its evidence.
How it works
Report
Paste the error, a screenshot or a plain description.
Investigate
AI searches code, logs, tickets and databases and links the evidence.
Explain
You get the likely cause, where it sits in the code and the evidence trail.
Built for teams with sensitive systems
Self-hosted
Runs on your own servers or private cloud, with local models if you need them.
Read-only access
Queries databases without changing data. Every query is logged.
Sensitive data masked
Define patterns for personal and card data before any AI analysis.
Visible reasoning
Watch what it searches and why, so engineers can check it.
Code-aware
Searches PHP, Java, C#, ASP.NET and Angular codebases.
Systems we work with
We connect to the platforms you already run. You keep your licences and your data.
- Jira
- Redmine
- ServiceNow
- GitHub
- GitLab
- Bitbucket
- SQL databases
- MySQL
- MongoDB
- Splunk
- Elasticsearch
Our own product, built in Australia
We built RootCauseAI because support teams lose hours switching between tools, and many cannot send code or logs to cloud AI. It runs inside your boundary, and your team decides what it can see.
- Your infrastructure, your access rules
- Never used to train a public model
- Architecture documents for your security review
Questions
How does RootCauseAI find the cause of an incident?
You describe the incident or paste the error, and RootCauseAI searches the relevant code, logs, tickets and databases, links the evidence together and reports the likely cause, where it sits in the code and why. Engineers see every step it took.
Does RootCauseAI send any data outside our network?
Not when it runs on local models: everything stays inside your environment. If you prefer to connect your organisation's approved model provider, you control exactly what is sent, and masking rules for personal and card data apply first.
Can RootCauseAI change production data?
No. Database connections are read-only, so it can query state but never write, update or delete records. Every query it runs is logged, which gives your security and audit teams a complete record of what it looked at.
What infrastructure does RootCauseAI need?
A server or virtual machine inside your own environment that can run containers. Local models run on standard hardware for most teams, and our engineers help your team install it, connect your tools and set access rules.
See RootCauseAI on a real incident
Book a walkthrough with our engineers.
Fixed-fee phases, agreed before we start. Stop after any phase.