# Edge Delta

> Edge Delta is a telemetry-native AI SRE: an immune system for your production.

Edge Delta is a telemetry-native AI SRE that continuously understands, investigates, and operates production. Most AI SREs begin work after an alert fires and query observability tools from the outside. Edge Delta's AI Teammates reason directly over the telemetry layer, on the live logs, metrics, traces, and events flowing through its Telemetry Pipelines, so they hold Production Context before an incident starts. The teammates run one loop on every incident: detect abnormal behavior, investigate across telemetry and change history, reason to a root cause backed by evidence, act through approval-gated Workflows, verify the fix against before-and-after telemetry, and learn by adding the incident to persistent Production Memory. The same Production Context serves SRE, security, and software engineering teammates, which is the path toward Autonomous Production.

## What is an AI SRE

An AI SRE is an AI agent that performs the work of a site reliability engineer. It detects abnormal behavior in production, investigates by gathering and correlating evidence across logs, metrics, traces, and deployment context, reasons its way to a root cause it can support with evidence, acts through approval-gated remediations, verifies that the fix worked, and learns from every incident. The category differs from earlier tooling in who runs the investigation: observability tools show humans what is happening, AIOps tools rank what humans should look at first, and an AI SRE performs the investigation itself and returns a conclusion a human can review.

## AI SRE

- [AI SRE](https://edgedelta.com/ai-sre): Edge Delta as a telemetry-native AI SRE: detection, investigation, root cause with evidence, and approval-gated remediation in one loop.
- [What is an AI SRE?](https://edgedelta.com/what-is-ai-sre): Definition of the AI SRE category, comparison with AIOps and observability, and ten evaluation criteria.

## Product

- [AI Teammates](https://edgedelta.com/product/agentic-teammates): The agents behind Edge Delta's AI SRE. Specialized teammates that detect, investigate, and remediate production incidents with evidence, approval-gated actions, and verified fixes.
- [Telemetry Pipelines](https://edgedelta.com/product/telemetry-pipelines): The telemetry-native architecture underneath the AI SRE. Controls and routes logs, metrics, traces, and events from any source to any destination.
- [Anomaly Detection](https://edgedelta.com/product/anomaly-detection): Detect and resolve incidents faster with correlated context across all services.
- [Security Data Pipelines](https://edgedelta.com/product/security-data-pipelines): Control and visibility over security data with real-time standardization, enrichment, and streaming.
- [Logs](https://edgedelta.com/product/logs): Process, transform, and enrich every log at the edge or in the cloud.
- [Metrics](https://edgedelta.com/product/metrics): Analyze metrics end-to-end to improve response times, lower costs, and mute noise.
- [Traces](https://edgedelta.com/product/traces): Deep visibility and full command over trace data with advanced tracing capabilities.
- [Integrations](https://edgedelta.com/product/integrations): 50+ integrations across analytics, alerting, storage, and more to connect Edge Delta with your existing tools.
- [Classic Observability](https://edgedelta.com/product/observability): Telemetry pipelines, logs, metrics, and traces on a distributed engine that scales without burning your budget.
- [Product Overview](https://edgedelta.com/product): Overview of the Edge Delta platform capabilities.

## AI Teammates

- [SRE Teammate](https://edgedelta.com/product/agentic-teammates/sre): The flagship AI SRE Teammate. Triages alerts, runs root cause analysis with supporting evidence, and recommends remediations.
- [Security Engineer Teammate](https://edgedelta.com/product/agentic-teammates/security-engineer): AI teammate for automated security monitoring and threat investigation.
- [Software Engineer Teammate](https://edgedelta.com/product/agentic-teammates/software-engineer): AI teammate that explores codebases, reviews pull requests, fixes bugs, and opens PRs, with every write gated on approval by default.
- [Work Tracker Teammate](https://edgedelta.com/product/agentic-teammates/work-tracker): AI teammate that watches alerts, security findings, and incidents, opens or updates tickets in Jira and Linear when work is missing or unowned, and links them back to the source.

## Use Cases

- [Use Cases Overview](https://edgedelta.com/use-cases): Explore how Edge Delta helps teams solve real-world observability challenges.
- [Agentic Incident Response](https://edgedelta.com/use-cases/ai-monitoring): AI Teammates triage alerts, run root cause analysis, and coordinate remediation automatically.
- [Kubernetes Observability](https://edgedelta.com/use-cases/monitor-kubernetes): Connect to any cluster and give AI Teammates clean, real-time signals to detect, triage, and fix issues before they reach customers.
- [Multi-Cloud Observability](https://edgedelta.com/use-cases/multi-cloud-observability): AI Teammates monitor production across private and public clouds and triage cross-cloud incidents.
- [Data Tiering](https://edgedelta.com/use-cases/data-tiering-strategy): Route high-value data to premium platforms, archive everything else, and give AI Teammates curated signal.
- [Unified Logging Architecture](https://edgedelta.com/use-cases/unified-logging-architecture): Collect, pre-process, and route all logs through a single pipeline that feeds every AI Teammate.
- [Avoid Vendor Lock-In](https://edgedelta.com/use-cases/avoid-vendor-lock-in): Telemetry Pipelines deliver data in any format, including OpenTelemetry, to any destination, with no lock-in to a single data or AI vendor.
- [Store Your Own Telemetry Data](https://edgedelta.com/use-cases/analyze-logs-within-your-environment): Keep logs, metrics, and traces in your own infrastructure while AI Teammates query them natively.

## Resources

- [Blog](https://edgedelta.com/company/blog/page/1): Latest posts on AI-powered observability, telemetry pipelines, and more.
- [Knowledge Center](https://edgedelta.com/company/knowledge-center/page/1): In-depth articles and guides on observability topics.
- [Videos](https://edgedelta.com/company/videos/page/1): Talks, interviews, and webinars.
- [Demos](https://edgedelta.com/company/demos/page/1): Product walkthroughs and recorded AI Teammates investigations.
- [Case Studies](https://edgedelta.com/company/case-studies): Customer success stories and real-world outcomes.
- [Events & Webinars](https://edgedelta.com/company/events): Upcoming events and webinars.
- [Newsletter](https://edgedelta.com/company/newsletter): Subscribe for product news, observability trends, and expert insights.
- [RSS Feed](https://edgedelta.com/feed.xml): Blog RSS feed.

## Compare

- [Edge Delta vs Datadog Bits AI](https://edgedelta.com/compare/datadog): How Edge Delta's telemetry-native AI SRE compares with Datadog Bits AI across detection, investigation, action, verification, learning, and pricing.
- [Claude + Grafana MCP vs AI SRE by Edge Delta](https://edgedelta.com/compare/claude-grafana-mcp): What a model gets from a query tool versus a production system: knowledge graph, guardrails, context, memory, orchestration, automation, and pricing.
- [Edge Delta vs Resolve AI](https://edgedelta.com/compare/resolve): How Edge Delta's telemetry-native AI SRE compares with Resolve AI across detection, investigation, action, verification, learning, and pricing.
- [Edge Delta vs Traversal](https://edgedelta.com/compare/traversal): How Edge Delta's telemetry-native AI SRE compares with Traversal across detection, investigation, action, verification, learning, and pricing.
- [Edge Delta vs Cribl](https://edgedelta.com/compare/cribl): Compare Edge Delta and Cribl across data processing, pricing, scalability, and automation.
- [Best AI SRE Tools in 2026](https://edgedelta.com/ai-sre-tools): A dated, sourced comparison of twelve established AI SRE tools by where an investigation starts, what the agent can see, how it acts, and what it costs.
- [Edge Delta vs Competitors](https://edgedelta.com/compare): How Edge Delta's telemetry-native AI SRE compares with the other agents in the category, with dated claims and public pricing.

## Benchmarks and Research

- [Benchmarks](https://edgedelta.com/benchmark): Open, neutral benchmarks for AI incident reasoning and observability pipeline performance, with public repos and dated leaderboards.
- [AI SRE Arena](https://edgedelta.com/arena): An open benchmark for AI SREs. Twenty-one Kubernetes faults injected into one shop, investigated by Edge Delta's and Grafana's native AI investigations and by two Claude control runs, and judged against the same answer keys under a published rubric. Edge Delta detected 18 of the 21 incidents on its own; Grafana detected 12.
- [RCA Bench](https://edgedelta.com/benchmark/rca-bench): Drops a frontier model into a frozen incident and asks which commit caused it. Leaderboard across frontier and open models.
- [Blast Radius Bench](https://edgedelta.com/benchmark/blast-radius-bench): When five services fail at once, can a model separate the root cause from the blast radius and recover the failure path?
- [Noise Bench](https://edgedelta.com/benchmark/noise-bench): Twenty alerts fire at 2am. Can a model decide who to wake up, catching every real incident without drowning in flaps?
- [Pipeline Performance Bench](https://edgedelta.com/benchmark/pipeline-performance-bench): HTTP log ingestion throughput compared across Edge Delta, Cribl, the OpenTelemetry Collector, and Fluentd under identical conditions.
- [Golden Threads Fine-Tune](https://edgedelta.com/benchmark/golden-threads-fine-tune): Case study in specializing a 32B model for incident diagnosis on real Edge Delta investigation threads, scored against its base and frontier models.
- [Same Bug, Four Minds](https://edgedelta.com/same-bug-four-minds): One production panic paged four frontier-model agent teams at once. All found the same root cause and diverged on how they proceeded.

## Demos

- [What is an AI Teammate?](https://edgedelta.com/company/demos/what-is-an-ai-teammate): Short explainer on AI Teammates and how they differ from chat assistants and AIOps.
- [SRE Teammate Walkthrough](https://edgedelta.com/company/demos/sre-teammate-walkthrough): The SRE Teammate triaging an alert, investigating, and reaching root cause with evidence.
- [Security Engineer Teammate Walkthrough](https://edgedelta.com/company/demos/security-engineer-teammate-walkthrough): The Security Engineer Teammate investigating a finding end to end.
- [Three Popular SRE Teammate Use Cases](https://edgedelta.com/company/demos/three-popular-sre-teammate-use-cases-for-observability-teams): Alert triage, root cause analysis, and remediation as observability teams run them today.
- [Introducing Edge Delta Workflows](https://edgedelta.com/company/demos/introducing-edge-delta-workflows): Approval-gated Workflows that let AI Teammates act on production safely.
- [Introducing Edge Delta's Knowledge Graph](https://edgedelta.com/company/demos/introducing-edge-deltas-knowledge-graph): The service and dependency graph AI Teammates reason over during investigations.
- [Trust AI Teammates in Production](https://edgedelta.com/company/demos/trust-edge-deltas-ai-teammates-in-production): Guardrails, evidence, and approvals that make autonomous investigation safe to run.
- [Resolve Production Issues Faster with the AI Team](https://edgedelta.com/company/demos/resolve-production-issues-faster-ai-team): Multiple AI Teammates coordinating on a single incident.
- [Frontier Model Race](https://edgedelta.com/company/demos/frontier-model-race-technical): Frontier models racing to diagnose the same incident on Edge Delta.
- [Observability for the AI Era](https://edgedelta.com/company/demos/observability-for-the-ai-era): Why AI SREs need a telemetry-native foundation rather than a query tool bolted onto dashboards.

## Customers

- [Box](https://edgedelta.com/company/case-studies/box): Box uses Edge Delta Telemetry Pipelines to manage log collection, processing, and routing across the organization, reducing cost and complexity.
- [Banco Inter](https://edgedelta.com/company/case-studies/inter): Banco Inter reduced observability costs by 60% with a distributed observability approach and gave each team control over its data.
- [Webscale Networks](https://edgedelta.com/company/case-studies/webscale): Webscale consolidated its observability stack on Edge Delta and cut debugging times by 91%.
- [Qualiti](https://edgedelta.com/company/case-studies/qualiti): Qualiti improved developer log search and time-to-triage by up to 70%.
- [Fama](https://edgedelta.com/company/case-studies/fama): Fama reduced Datadog costs by 80% and automated anomaly detection.
- [MediaKind](https://edgedelta.com/company/case-studies/mediakind): MediaKind optimized log data by over 80% and derives insights upstream.
- [Super League Gaming](https://edgedelta.com/company/case-studies/super-league-gaming): Super League Gaming analyzes 100% of its log data and sped up root cause identification.
- [Peerspace](https://edgedelta.com/company/case-studies/peerspace): Peerspace automated monitoring and troubleshoots issues faster with Edge Delta.
- [Agi](https://edgedelta.com/company/case-studies/agi): Agi sets up alerts in minutes, anticipates production issues, and reduced observability costs.

## Developers

- [Developer Hub](https://edgedelta.com/developers): API, OpenAPI spec, MCP server, and CLI resources in one place.
- [Documentation](https://docs.edgedelta.com): Technical documentation, setup guides, and API reference for Edge Delta.
- [API Reference](https://docs.edgedelta.com/api/): REST API documentation for the Edge Delta platform (api.edgedelta.com).
- [OpenAPI Specification](https://edgedelta.com/openapi.json): Machine-readable OpenAPI 3.0 spec for the Edge Delta API.
- [API Catalog](https://edgedelta.com/.well-known/api-catalog): RFC 9727 catalog linking the API spec and docs.
- [API Authentication](https://docs.edgedelta.com/create-api-token/): How to create and use Edge Delta API tokens.
- [MCP Server](https://docs.edgedelta.com/edge-delta-mcp-server/): Hosted Model Context Protocol server at https://api.edgedelta.com/mcp (streamable HTTP) for AI agents like Claude, Cursor, and VS Code. Manifest: https://edgedelta.com/.well-known/mcp
- [edx CLI](https://github.com/edgedelta/edx): Official Edge Delta command line interface. Install with `brew install edgedelta/tap/edx` or `go install github.com/edgedelta/edx@latest`.
- [GitHub](https://github.com/edgedelta): Edge Delta open source repositories, including the agent, MCP server, and CLI.

## Company

- [About Us](https://edgedelta.com/company/about): Our story and mission.
- [Careers](https://edgedelta.com/company/careers): Join Edge Delta and help shape the future of observability.
- [Partners](https://edgedelta.com/partners): Edge Delta Partner Program for strategic growth.

## Getting Started

- [Start Free Trial](https://edgedelta.com/start-free-trial): 14-day free trial of AI Teammates.
- [Pricing](https://edgedelta.com/pricing): AI Teammates pricing information.
- [Documentation](https://docs.edgedelta.com): Technical documentation and guides.

## Legal

- [Privacy Policy](https://edgedelta.com/privacy-policy)
- [Terms of Service](https://edgedelta.com/terms-of-service)
- [Free-Trial Subscription Agreement](https://edgedelta.com/free-trial-agreement)
- [Subprocessors](https://edgedelta.com/subprocessors)
- [Do Not Sell / Share My Info](https://edgedelta.com/do-not-sell-share-my-info)
- [GDPR Privacy Notice](https://edgedelta.com/gdpr)
- [Privacy Shield Notice](https://edgedelta.com/privacy-shield)

## Sitemap

- [XML Sitemap](https://edgedelta.com/sitemap.xml)