What Is Self-Driving Production? The Future of AI SRE Explained
Traversal CEO Anish Agarwal joins EMA’s Dan Twing on Tech Transformed. They discuss why AI-written code makes production incidents harder to fix, and how enterprises move from AI suggestions to autonomous remediation.
00:00
/
00:00
A ServiceNow ticket comes in at a large bank: 50 customers can’t log in. Between 60 and 100 people sit in a war room for two hours before they find the cause, a slow SQL query ten hops down the stack. This is the problem Traversal was built to solve.
On the Tech Transformed podcast, Traversal CEO Anish Agarwal joins EMA’s Dan Twing to explain how Traversal helps enterprises move toward self-driving production.
What they cover
- Why AI-written code makes incidents harder to troubleshoot
- How Traversal finds root causes across thousands of apps
- How teams move from AI suggestions to autonomous remediation
How Traversal finds the root cause
AI-generated code can run without errors and still give the wrong answer, and most failures happen where systems connect. Dashboards weren’t built for that. Traversal takes a different approach:
- A production world model maps entities, their connections and live telemetry, so agents can reason over all of production.
- Causal analysis separates real upstream and downstream anomalies from noise, even when many alarms fire at once.
- Converging evidence builds confidence: when several independent paths point to the same cause, that’s the root cause.
