Welcome to the DevNet Month of Cisco Validated Designs
Cisco Validated Designs give you tested, supported blueprints, and AI design review catches problems before anyone pulls a cable. Watch episode 1 and join us every Thursday.
Cisco DevNet
Mastering stateful stream reordering at 200,000 messages per second
Our Flink job hit a wall at 200,000 messages per second because of RocksDB. Here is the in-memory reordering pattern that fixed it, and how A/B development proved it safe.
Design-to-Code: Reimagine Handoff Workflow with AI – Rebuilding PubHub Part 3
How the PubHub team replaced the Figma handoff with an AI-built, designer-refined HTML prototype, and what worked and didn't along the way, so other teams can adapt it.
Policy as Code for the Network: Guardrails with Open Policy Agent
Learn how to use Open Policy Agent (OPA) to enforce network guardrails as code. Catch insecure configurations before deployment and maintain a unified security posture across your entire enterprise network.
Scan Code for Vulnerabilities with Cisco Antares in DevNet Learning Labs
Two GPU-backed Antares Learning Labs are live. Use a small open-weight model to narrow a repo to the files that may hold a CWE, then see how DevNet shares one GPU across learners.
Using AI to re-architect the Cisco DevNet sandbox environment – Part 1
An AI agent reviewed our sandbox network design, caught routers hanging off the spines, and asked one question that turned a Clos fabric into a two-tier vPC design.
Antares: a Security Analyst That Never Leaves Your Laptop – Security Loop Series, Part 1
A CWE advisory names your dependency, but not where the bug lives. See how Cisco's Antares-1B localizes it offline on a laptop, with auditable traces and honest benchmark numbers.
Three languages, one binary: Building a VMware Nutanix inventory with Claude
Two UIs, two CSV shapes, and a spreadsheet I kept rebuilding. Claude helped turn throwaway scripts into one Go binary that inventories VMware and Nutanix in a single run.
The receipt, not the demo: Four questions that earn AI agent trust
A good demo proves nothing. Four questions about policy, approvals, and audit logs show whether an AI agent has earned the right to touch production.