As I mentioned in my previous blog on the CCNP Automation core exam (AUTOCOR), I wanted to provide an insider's view of the exam topics for the new professional automation track. I consider myself an insider because I was part of the team of product managers, psychometricians, program managers, and subject matter experts that helped discuss (and sometimes argue) for what should and sshouldn'tbe included in the certification.
This culminated in a lot of hours locked in a room to hash out the exam topics' lists that are now available for review. To say that it is a lot of exhausting work is an understatement. Still, the team's entire focus is to deliver a comprehensive exam that you, the learner, find valuable enough to pursue in your professional career.
However, I know that there are sometimes questions as to why things were built or tasked a certain way, and its (E)-NAUTO problem (not a' problem, get it?|?YouTube) for me to take some time out of my day to answer your burning enterprise automation exam topic questions.
ENAUTO 1.1 Removed Topics
The first big departure is from the previous ENAUTO 1.1 domain 1. This domain focused on software development practices. While topics like git usage and functionality are important, most of the domain dealt with proper coding and development practices, interpretations of Python scripts, and the benefits of automation tooling (virtual environments, infrastructure as code, and the like). We wanted to evolve the ENAUTO to focus more on the functionality and practitioner level. We took those concepts and either removed them completely (do we need to know the benefits of Infrastructure as Code or should we be able to utilize IaC tooling in an automation solution?) or incorporated them into a larger, automation focused task. Understanding these topics is still valuable. However, a network automation engineer in today's world doesn't need to sell the benefits of APIs; instead, they are being asked how to deliver the greatest value to an organization by leveraging existing tools, minimizing disruption, and avoiding unnecessary tool-spread.
Like the AUTOCOR exam, the introductory paragraph for the exam topics specifically calls out certain platforms and operating systems. However, you might notice a separation between devices and platforms within the exam domains-an intentional change. Every enterprise has unique needs. For some, managing all devices through a controller or platform makes sense; others may prefer device-by-device management using automation and orchestration tooling. Both skills hold importance, so both domains have been included in the exam, with each focusing on its respective area.
The device configuration domain is Python heavy, focusing on a variety of libraries and protocols that are commonly used to manage devices.
Controllers and platforms, however, are also important because they provide easy automation and single "pane" visibility across large, geographically separated, disparate enterprises. These platforms are also developed with APIs at their core, allowing for functionality, configuration, and observability to be extracted via programmatic methods, a win for automation developers!
While there is some overlap, the intent is to ensure that the capabilities provided by the platforms, like additional Jinja2 templating or creating larger automation routines leveraging the APIs provided by the platform (e.g. kicking off a device upgrade routine using an API, rather than building the automation yourself and using that as part of an onboarding routine). This isn't to push or "sell" folks on one management style vs. another, just to illustrate the common tasks engineers face in both scenarios.
Keen eyes will notice the software platforms in scope at the top of the exam topics sheet. Even keener eyes may see that there are two platforms that are completely new to the automation track, Cisco Identity Services Engine (ISE) and Cisco ThousandEyes. These may seem out of place, but let's dive into why these platforms are important.
Cisco ISE provides a foundational "single source" for all device access and user identity within an enterprise network.
Let's keep in mind, this isnota security exam; learners will not be expected to configure ISE, but it is in scope to be able to use ISE APIs as part of a segmentation or policy enforcement automation within the campus switches and routers.
Cisco ThousandEyes operates at the other end of the deployment, not in provisioning, but in operations and management. End-users of the network care about the quality of their experience within the network; they don't care whether the configuration applied to their switch was automated or typed by hand on the CLI, it just needs to work.
This data could then drive a specific outcome to the automation (rollbacks, anyone?), which would also be automated. Again, you don't need to create the tests but know how to interpret the data and results from tests as a pre- and post-change validation to your automation.
Speaking of pre- and post-change, have you heard about Cisco Modeling Labs? If you haven't, it's a network simulation platform that runs Cisco Virtual Network Functions (VNFs), like virtual routers, switches, firewalls, and wireless controllers.
These VNFs can be connected to each other in a virtual canvas, which allow a user to model a network, its configurations, and the behavior because of those changes. It provides full control-plane simulation if routes are changed, VLANs are added, or ACL rules are removed. Network engineers are using it to study for a variety of exams from associate to expert because of the capabilities in the platform.
Building Digital Twins using Cisco Modeling LabsThe fusion was natural: network managers not trusting the automation to do what it said it would do; network automation engineers wanting to study and prove that their automation solutions worked. Cisco Modeling Labs fills that need for pre-change testbed which becomes the post-change standard for all configurations. Network automation engineers should be fluent in Cisco Modeling Labs and the idea of digital twins as a testbed for their automation before implementation into production. We all like to avoid those "resume generating events", right?
I know, I know. AI is everywhere, and you may or may not be facing AI burnout. However, as someone who has always viewed practicality first and foremost when learning something new, you can rest easy that this was put into the exam topics for reasons beyond just making sure that we "checked the AI box".
This task is intentionally written in the same verbiage as that from the AUTOCOR exam topics. As an automation professional, we foresee that it will be your responsibility to translate tools and capabilities into something that can be used by AI for automation routines. Being able to design, develop, and scope the configuration for that MCP appropriately will be a very critical skill to have, which is why it appears in both AUTOCOR and ENAUTO (and I bet you can guess what other exam has it as a task as well).
Through this major uplift on the exam, we hope that the skills that are being asked are reflective of common problems that today's enterprise automation engineers are being asked to solve. We tried to strike the right mix of controllers and platforms, but also tie in validation and testing, because those are important concepts to include in your workflow, and we hope that all these skills can be used in your professional career.
As always, you can find me on LinkedIn or on X (@qsnyder) if you have any questions regarding this or any other exam. Happy studying!
Sign up for Cisco U. | Join the? Cisco Learning Network?today for free.| Join the? Cisco Learning Network?today for free.
Learn with Cisco
X?|?Threads | Facebook?|?LinkedIn?|?Instagram|?Threads | Facebook?|?LinkedIn?|?Instagram?|?YouTube
Use? #CiscoU and #CiscoCert?to join the conversation.
CCNP Automation: A Renamed Certification, Reimagined
Learn with Cisco: Evolving for the Age of AI, Automation, and Cloud