JOHN CAPOBIANCO / START HERE

Learn network automation.
Build something useful.

Choose the problem you want to solve, then follow a path through my original articles, code, books and interactive examples.

One small lab. One visible result.

Start with data you can inspect and a result you can explain. The articles retain their original publication dates and historical examples; check current project documentation and software compatibility before reproducing a setup.

Meet John: author, engineer and Google Developer Expert →
LEARNING PATH

Python network automation with pyATS and Genie

Start with a network command, turn its output into structured data, and use that data to test an expectation. This path connects the early Python experiments to the Cisco Press book and today's MCP work.

Before you start

Basic Python, a lab device you can access, and familiarity with its show commands.

Try this next

Choose one interface or routing check. Save the parsed result, state what healthy looks like, then repeat the check after a controlled lab change.

  1. My First Pure-Python Network Automation with pyATS / Genie !Published 2021-03-12 · 5 min read
  2. Letting the Genie out of the bottlePublished 2020-11-22 · 8 min read
  3. Test-Driven AutomationPublished 2022-10-14 · 1 min read
Explore the pyATS book, code and videos →
LEARNING PATH

Network visualization: CLI data to mind maps and 3D

Make network state easier to explore. Begin with a small, structured dataset, then compare a mind map with a 3D representation. The interactive visualization lab lets you inspect the same fictional interface data in several forms.

Before you start

Comfort reading interface tables or JSON. The browser lab needs no network credentials.

Try this next

Follow one interface from the CLI table into the mind map and 3D scene. Toggle the lab's link-down example and compare how each view communicates the change.

  1. Introducing Mind nMap – An Open Source Network Mind Map ToolPublished 2021-12-16 · 1 min read
  2. Show IP Interface Brief – Reimagining in 3DPublished 2021-08-07 · 1 min read
  3. Learn how to make “business ready” documents!Published 2022-04-05 · 1 min read
Try the mind map and 3D network lab →
LEARNING PATH

NetClaw and AI agents for network engineering

Explore NetClaw through its development journals, network integrations and security work. Pair the experiments with the current repository documentation; capabilities and setup evolve faster than an archived article.

Before you start

An isolated lab, a clear understanding of the tools an agent can call, and access limited to the work you intend it to perform.

Try this next

Trace one task from the user's request to the tool result. Start with observation, inspect the evidence, and define an approval point before a tool can change network state.

  1. 6 days of netclaw – by netclawPublished 2026-02-27 · 10 min read
  2. NetClaw + IP Fabric: Network Assurance Meets AI-Powered AutomationPublished 2026-06-27 · 3 min read
  3. Introducing DefenseClaw: Enterprise Security for NetClawPublished 2026-04-18 · 2 min read
Explore NetClaw, NCFED and the mobile app →
LEARNING PATH

Ansible network automation and documentation

Work through facts, loops and report generation. These original field notes show the progression from collecting network data to producing documentation that other people can use.

Before you start

YAML fundamentals, an Ansible inventory, and a small lab with known expected results.

Try this next

Collect a small set of facts and render a report. Measure collection and rendering separately before changing the playbook, so a faster result is still a correct result.

  1. Just the facts ma’amPublished 2020-11-18 · 17 min read
  2. Dynamic Loops in AnsiblePublished 2020-12-01 · 5 min read
  3. Ansible Performance – Moving to Jinja2 for Automated DocumentationPublished 2020-12-02 · 4 min read
Read the complete Automate Your Network book →
LEARNING PATH

Cisco ISE automation: ERS and MnT APIs

Two complementary parts of the Cisco ISE archive: External RESTful Services and the Monitoring and Troubleshooting API. Read the original examples alongside the documentation for your installed ISE release.

Before you start

An ISE lab, API access appropriate to the endpoint, and a way to inspect HTTPS responses without exposing credentials.

Try this next

Identify whether the question concerns configuration or operational state, choose the appropriate API, and inspect a read-only response before building a workflow around it.

  1. By Popular Demand – Automating the Cisco Identity Services Engine (ISE) External RESTful Services (ERS) API SuitePublished 2021-02-20 · 7 min read
  2. Automating the Cisco Identity Services Engine Monitoring and Troubleshooting Node (ISE MnT) REST APIPublished 2021-02-17 · 10 min read
Browse the complete technical archive →
LEARNING PATH

Local AI, RAFT and network knowledge

Explore experiments with domain-specific model knowledge, local inference and agents. Keep the distinction between a model's answer and verified network evidence visible throughout the workflow.

Before you start

Python familiarity, the hardware requirements of the selected model, and a small dataset you are allowed to use.

Try this next

Keep a set of questions outside the training data. Compare answers against source material and record failures as carefully as successes before putting the model into a network workflow.

  1. Augmenting Network Engineering with RAFTPublished 2024-04-27 · 7 min read
  2. Local RAFT: Fine-tuning Llama3 with domain-specific knowledge locally and privatelyPublished 2024-05-11 · 38 min read
  3. Running Cisco Antares-1B Locally (macOS / Apple Silicon)Published 2026-08-11 · 8 min read
Explore the Google Developer Expert collection →
THE BOOK / AN INTERACTIVE COURSE

Automate Your Network on Educative.

Prefer a guided course? The Educative course built around my first book covers Ansible, network reconnaissance, repository structure, Jinja templates, configuration management and CI/CD.

Course access and current enrollment terms are managed by Educative. The full book remains free to read here.

Keep learning with the community.

Watch a build, explore the repositories, or join a conversation with engineers working through the same questions.