superlog

Superlog is an observability product for teams running cloud software. Instead of asking you to hand-wire every metric, log, and trace across dozens of services, it instruments your code automatically and builds the observability dashboards and alerts from what it finds, so setup runs once and stays useful. That's the pitch.

The problem it targets is drift. Dashboards go stale. Alerts fire on services that no longer exist, and someone has to spend a Friday afternoon cleaning up after every deploy because nobody else wanted to own that chore. Superlog does root cause analysis with help from an agent, so the first pass of an investigation isn't left to a tired human. It leans on OpenTelemetry, the open standard for telemetry data, so the signals it collects aren't locked to one vendor. According to the company, that agent also handles part of the incident work itself: it investigates what broke, works out why, and can ship a fix as a pull request for your team to review.

The main limitation to know up front is scope. This is tooling for people who ship and run software, not a general analytics product. If your team doesn't already run cloud services or have anywhere to review pull requests, most of it won't apply to you. There's also less independent testing and review out there than you'd find for older monitoring incumbents. So it pays to kick the tires yourself before you commit a whole on-call rotation to it.

Getting Started

1. Create an account on the Superlog site and connect the repository or environment you want to observe.
2. Let the auto-instrumentation hook into your code so traces, metrics, and logs start flowing through OpenTelemetry.
3. Review the dashboards and alerts Superlog generates, and trim anything your team doesn't care about.
4. Point the incident agent at your services so it can investigate failures and draft fix pull requests.
5. Wire notifications into your existing channels so alerts reach the people on call.