NOTE: Demo visuals use either blurred real data or synthetic placeholders to protect customer privacy.
The Drag of Manual Triage
Before Nora, every alert started with the same mechanical sequence. The engineer checked the runbook, opened Prometheus or Grafana, pulled logs from CloudWatch, and finally hunted for recent code changes. Each step added friction, especially overnight when attention and patience were low.
- Searching multiple systems for relevant data.
- Comparing metrics across dashboards manually.
- Guessing which recent deploys might be the cause.
Those steps slowed incident response and increased the risk of errors. The team needed a way to get reliable context fast.
Nora’s First Version Takes Shape
Rishi designed Nora as a Slack-based responder that reacts the instant a PagerDuty alert hits. Within seconds, Nora gathers evidence and posts a triage summary directly in the alerts channel.
- Runs Prometheus queries to surface metric anomalies.
- Pulls related runbook steps from Confluence.
- Scans recent PRs to identify likely code changes.
- Posts next-step recommendations and links in Slack.
By doing this automatically, Nora shifted the engineer’s role from searching to acting. Each triage now begins with one post that consolidates everything needed for a response.
Early impact on engineering time
The effect was immediate. Teams saw faster triage and smoother handoffs between time zones. Engineers could skip redundant checks and trust Nora’s initial assessment before diving deeper.
- Response time for alerts improved across multiple teams.
- Context switching dropped, cutting confusion during incidents.
- Engineers reclaimed hours previously lost to manual data gathering.
In the next phase, Rishi aims to expand Nora’s access to logs and external systems to enrich and improve the reliability of the summaries.
Early Reactions from Peers
Colleagues quickly saw the difference between Nora and commercial options. It aligned with internal workflows, adhered to Slack norms, and shipped in under a week. That combination led the team to lean toward building rather than buying. The speed and precision of early results set the stage for broader adoption.
