The AI SRE that catches issues before you get paged

Herald detects potential issues, investigates them across code, infrastructure, and telemetry, and hands you the root cause — all before an alert fires or a customer complains.

70%+ accuracy on novel incidents

1. Anomaly detected error rate

• HTTP 500 spike on Slack event ingestion pipeline

2. Anomaly detected stack trace 5s later

• Outbound timeouts in slack_bolt → aiohttp auth path

3. Cluster grouped 30s later

• 2 related error signals correlated across alert pipelines

4. Investigating 10s later

• Correlating error groups across ingestion and alert pipelines
  • Tracing auth.test call in Slack Bolt authorize hot path
  • Matching prior aiohttp → slack_sdk timeout pattern

5. Root cause determined 2m later

• auth.test API call timing out during authorization, causing intermittent 500s on the ingestion hot path.

• Cache AuthorizeResult · configure explicit SDK timeouts

Trusted in Production

The metric stays below the alert threshold for most of the chart, then crosses it.

Marker 1 identifies the threshold crossing. The area above the threshold after that crossing is labeled no runbook coverage, and marker 2 identifies the uncovered alert territory.

api latency

time

1 alert threshold

2 no runbook coverage

1 threshold model

Others Require Alert Thresholds

You have to instrument, tune thresholds for each data stream, and anticipate every failure mode worth watching. Miss one, and you're blind to it.

Others Require Runbooks

You document every investigation workflow before it's needed. Maintain them as your stack evolves. When something novel breaks, there's no runbook and no investigation.

Other agents only handle failures someone already documented. Herald investigates the ones nobody saw coming. No runbook required.

Learn

Herald builds a context graph before any alerts fire — observability, codebase, CI/CD, docs, and dependencies — so it knows what normal looks like and can work to solve any problem.

context: Jira tool 65 · config read path · CUST-8291-X

Detect

No thresholds to set. Herald builds a custom anomaly detection model for each data stream and surfaces validated issues before your customers notice.

validated signal: HTTP 500s rose to 18–26% over 22 minutes while other tenants stayed flat

Investigate

Never write another runbook. Herald evaluates multiple hypotheses simultaneously, each against the right data source, and delivers RCAs in minutes.

RCA: Schema Drift · legacy Vault keys rejected after PR #4275

Herald's agent onboards and adapts quickly. Gartner's 2026 AI SRE Market Guide identifies proactive incident prevention and contextual awareness as next-generation capabilities. Herald already does both.

• Results in days, not months. The Herald agent learns your stack quickly and efficiently – see your first RCA in days.
• Solves the unknown. 70%+ accuracy on novel incidents for one of the world's biggest B2B2C platforms.
• Never repeats mistakes. Herald learns from every single investigation, so it never makes the same mistake twice.

Powered by UC Berkeley research

Herald was founded by PhDs and Professors from UC Berkeley's innovation center, RISELab, combining expertise in AI, LLMs, data systems, and scalable infrastructure.

The Herald Approach

1. Up and running in minutes Herald scans your codebase, finds what tools you use, and configures itself instantly.
2. Grounded in your product Context graph ensures grounded, token-efficient work.
3. Built for engineers 100% RCA accuracy on known issues; 70% on novel ones.

Predictive detection. Novel incidents. No runbooks.

Read the latest

What It Actually Takes for AI to Map Your Production Systems
An agent cannot diagnose incidents without understanding your system and how it's connected. Here is what that involves.

System Understanding Is the Whole Game for AI SREs
Momento CTO Daniela Miao on why an AI SRE can only be trusted with alerting decisions if it understands how a system is built.

Engineering Knowledge Replicants
Tribal knowledge isn't collective. It's fragmented and fragile, and incident response depends on it. Why documentation and DIY agents still fall short for AI SREs.