Sonarly connects to observability platforms (e.g., Sentry, Datadog, PagerDuty) and code repositories to collect full telemetry bundles—logs, traces, metrics, and code diffs—and runs Claude Opus 4.6‑based root‑cause inference. The platform clusters duplicate alerts, posts concise analysis to Slack or Discord, and automatically creates pull requests with suggested code changes for remediation.
Funding
Funding not disclosed
Founders
Product
Problem
Engineering teams receive high‑volume alerts from observability platforms but must manually aggregate logs, traces, and metrics to diagnose incidents. Duplicate notifications and fragmented context force developers to switch between tools, extending mean time to resolution and increasing operational overhead.
Solution
Sonarly acts as an autonomous incident‑triage layer that connects to existing observability stacks (e.g., Sentry, Datadog, PagerDuty) and source‑code repositories. The platform ingests each alert with its full telemetry bundle—logs, distributed traces, performance metrics, and relevant code snapshots—and feeds it to a large‑language‑model (Claude Opus 4.6) for root‑cause inference. AI‑driven clustering collapses duplicate alerts into a single investigation, while the system posts concise analysis directly into the team’s Slack or Discord channel. When a fix is identified, Sonarly automatically generates a pull request against the linked GitHub repository, streamlining remediation without leaving the communication workflow.
Target Audience
The primary customers are DevOps, Site Reliability Engineering, and software development teams that operate cloud‑native, micro‑service architectures and rely on continuous observability pipelines.
Features
- Bi‑directional connectors for major observability tools (Sentry, Datadog, PagerDuty) and version‑control systems (GitHub, GitLab)
- Full‑stack telemetry aggregation delivering logs, traces, metrics, code diffs, and historical incident data per alert
- Claude Opus 4.6 powered root‑cause analysis that queries logs and traces across micro‑service boundaries
- AI‑based duplicate‑alert clustering to eliminate redundant investigations and reduce token consumption
- In‑channel reporting: threaded messages with actionable insights posted to Slack or Discord, eliminating context switching
- Automated pull‑request generation with suggested code changes, ready for reviewer approval
- Enterprise‑grade security features including SOC 2 compliance, SSO integration, end‑to‑end encryption, and role‑based access controls