The On-call AI platform that puts AI to work on your production issues.

tl;dr

Connect Doctor Droid to your alerting & monitoring stack. Here’s what you get:

1. Doctor Droid can auto-initiate investigation (from an alert in Slack), automatically fetching relevant information from different tools (log search, metrics, recent deployments, related commits, related alerts).

   1. Fetch additional data by replying in the thread (e.g. “@Doctor Droid can you look for deployment for service x?”)

   

   Tag Doctor Droid Slack bot in alert’s thread as a reply and get your investigation thread

2. When you are reported an issue, manually initiate an investigation

1. Get recommendations on every alert investigation basis the following:
   1. Past investigations on the platform
   2. On-call SOPs (Droid has integrations with Confluence, Notion & GDrive)
   3. Existing Playbooks in the platform
   4. Subject Matter Expertise, contextual to your company (currently can auto-investigate for issues/alerts related to code errors, golden signals, kafka, postgres, mongodb, hosts, redis)

Unified Investigation Platform

Doctor Droid brings together tools, people and document knowledge in the same place to accelerate investigation of issues.

Collaboration:

Run a collaborative investigation where users don’t have to share screenshots or lose information in Slack chats.

• A single investigation can have multiple users fetching data.
• Sharing an existing investigation (link) will bring the other developers within the team up to speed.

Data Sources:

Type 1: Read/Fetch (Investigations):

1. Metrics: Datadog, New Relic, Cloudwatch, Prometheus (via Grafana), Prometheus (direct), Mimir, Google Cloud Monitor, Azure Monitoring, Victoria Metrics