Recipes for Automation - A Look Inside Eric Chou's AI Networking Cookbook

March 20, 2026·1 hr 2 min·5 clips

Eric says an LLM cannot reason, so prompts must supply context, examples, and a clear format.

Eric Cho joins Heavy Networking to discuss his AI Networking Cookbook. Ethan Banks and Drew Connery-Murray frame the conversation around what it means to add artificial intelligence to the network automation toolbox. The first concern is context. Eric explains that early chatbot experiences were limited because you had to paste in nearly everything you wanted the model to know, whether that was a router config, a question about a line in the config, or another network task. That burned through the context window fast. The conversation then moves to the next step in the evolution of these tools: system messages. Those let you keep persistent instructions with the chat, so you do not have to restate your role, preferences, or goals every time you start over. The point is not magic. It is about giving the model enough information to answer more accurately and more usefully. Eric and the hosts keep coming back to the operator mindset. If you are asking an AI to help with network work, you should tell it exactly what you want, what format you want it in, and what kind of network engineer perspective it should assume. That is why examples matter. The episode keeps its footing in practical configs and prompts rather than drifting into a generic AI trend story. The tone stays grounded. The discussion sounds like engineers trying to make a new tool fit into real workflows instead of pretending the workflows do not matter. Because Eric also hosts Network Automation Nerds, the conversation lands naturally in the overlap between networking practice and automation habits. The book title comes up as part of that broader push to make AI useful to people who already live in the weeds of configs and operational details. The hosts keep the focus on what a network team would actually do with this kind of tooling. By the end, the takeaway is clear: better results come from better context, clearer instructions, and a more deliberate prompt. Listeners who are exploring AI for network automation will find the most value here.

As heard by us

A practical take on AI for network automation, with context and prompting doing the real work.

Heavy Networking uses Eric Chou's AI Networking Cookbook to keep the focus on practice, not hype. It treats AI as a tool for network automation that works better when prompts carry context, role, and clear instructions.

Read the full review in PlayNext →

Why you'd press play

You want AI prompts that behave like operator instructions for network automation, not one-off chatbot guesses.

Read the full recommendation in PlayNext →

Listen to the show on

Apple Podcasts Spotify

Play clips from this episode More from Heavy Networking