The Everything Feed - All Packet Pushers Pods HN812: Nokia EDA: AI Ops You Can Trust (Sponsored)
Jan 30, 2026
Stephen Butler, product line manager at Nokia for Event-Driven Automation, explains how EDA provides multi-vendor abstractions, schema-driven models, and programmable workflows. He discusses combining LLMs with deterministic tools, digital twin simulation for CI/CD validation, and an extensible app model for safe, testable AIOps.
AI Snips
Chapters
Transcript
Episode notes
Abstract Once, Manage Many Vendors
- EDA uses multi-vendor abstractions so users don't need to learn each device OS's syntax.
- The platform normalizes intent and operational state so automation sees the same model across vendors.
Pair Intent With Status
- EDA models resources with a spec (intent) and status (operational state) to combine config and telemetry.
- That pairing lets the platform detect configuration drift and produce higher-level health scores.
Give AI Structured Context
- Provide the LLM with the right context and pre-processed resources rather than raw device dumps.
- Use normalized, combined config-and-state models so AI can reason reliably about health and root cause.