Skip to content
Log in Sign up My account

ENCOR 6.0 Automation and Artificial Intelligence (15%)

Automation is mostly a read and interpret domain. You will be shown Python scripts, JSON bodies, YANG trees, API exchanges and EEM applets, and asked what they do or what is wrong with them. Only two objectives ask you to produce something: construct valid JSON (6.2) and construct an EEM applet (6.6). At 15% of the exam this is worth as much as Architecture, and it rewards practice more than memorisation: an hour of running scripts and curl commands against a Catalyst 8000v teaches more than a day of reading.

Objective Verb Page
6.1 Basic Python components and scripts Interpret Python and JSON
6.2 Valid JSON-encoded files Construct Python and JSON
6.3 High-level principles and benefits of a data modeling language, such as YANG Describe YANG Data Modeling
6.4 APIs for Cisco Catalyst Center and SD-WAN Manager Describe Catalyst Center and SD-WAN Manager APIs
6.5 REST API response codes and results in payload using Catalyst Center and RESTCONF Interpret REST Response Codes and Payloads
6.6 EEM applet to automate configuration, troubleshooting, or data collection Construct EEM Applets
6.7 Agent vs. agentless orchestration tools, such as Chef, Puppet, Ansible, and SaltStack Compare Orchestration Tools

Time budget: week 10 of the study plan. A workable split for 8 to 10 hours:

Days Pages Lab
1 to 2 Python and JSON, YANG Parse RESTCONF output with Python; explore models with NETCONF and RESTCONF
3 to 4 Controller APIs, REST response codes DevNet sandbox calls; provoke every status code on R1
5 EEM Four applets, including the data-capture applet
6 Orchestration tools, AI context section below Idempotent Ansible playbook
7 Review Practice questions and Sim 12

All labs share one small topology: an Ubuntu automation host (AUTO, 192.168.100.50) and a Catalyst 8000v (R1, 192.168.100.1) on the management subnet, plus R2 for the EEM lab. IOSv does not support NETCONF or RESTCONF, so use the Catalyst 8000v for those labs. See Lab Setup.

graph LR
  AUTO["AUTO Ubuntu<br/>Python, curl, Ansible<br/>192.168.100.50"] -- "192.168.100.0/24" --- R1["R1 C8000v<br/>RESTCONF, NETCONF, EEM<br/>192.168.100.1"]
  R1 -- "Gi2 10.1.2.0/30" --- R2["R2<br/>EEM lab peer"]
  AUTO -. "HTTPS to DevNet sandboxes" .-> SB["Catalyst Center and<br/>SD-WAN Manager"]

Related material:

AI and machine learning in network operations (context only)

Not a listed sub-objective

The v1.2 blueprint renamed this domain "Automation and Artificial Intelligence", but none of the sub-objectives 6.1 to 6.7 mentions AI. This section is background so the term is not a surprise. Study 6.1 to 6.7 first, and do not expect detailed AI questions.

Cisco uses AI and machine learning mainly in assurance: turning large volumes of telemetry into a short list of real problems.

Capability What it does Where
Dynamic baselines Learns what is normal for a network, site or device over time (for example onboarding times or throughput), instead of relying on fixed thresholds Catalyst Center AI Network Analytics (a cloud-based ML service)
Anomaly detection Raises an issue when behaviour deviates from the learned baseline, which catches problems that a static threshold would miss or would flag too often Catalyst Center assurance
Root-cause reasoning Runs automated troubleshooting workflows that collect data and reason towards a likely cause Catalyst Center Machine Reasoning Engine
Endpoint classification Groups and profiles endpoints from their traffic so policy can be applied to them Catalyst Center AI Endpoint Analytics
Path forecasting Predicts WAN path quality from history and suggests better paths SD-WAN analytics (WAN Insights)
Generative AI assistants Natural-language questions and summaries over controller data Cisco AI Assistant in Cisco management platforms 
graph LR
  T["Telemetry<br/>NetFlow, syslog, SNMP,<br/>streaming, client data"] --> B["ML baseline<br/>what is normal here"]
  B --> A["Anomaly detection"]
  A --> I["Prioritised issue<br/>with likely cause"]
  I --> R["Suggested action<br/>or automated workflow"]
  R -. "engineer approves" .-> N["Network change"]

Points worth knowing at the level of general understanding:

  • ML needs good data. The assurance features above depend on the same telemetry this course covers in domain 4 (NetFlow, syslog, SNMP, streaming telemetry).
  • Baselines beat static thresholds where "normal" changes by time of day or by site, which is the main argument for ML-based assurance.
  • AI output is a recommendation. Generated configuration or a suggested fix should be reviewed, tested and applied through the same change control as any other change, and data sent to cloud AI services is subject to privacy and data-handling policy.
  • The automation skills in this domain (APIs, structured data, idempotent tools) are what let an AI-suggested action be applied safely and repeatably.

Feature names and packaging change between Catalyst Center releases. Check the current Catalyst Center documentation on cisco.com before relying on any specific feature name.

When you finish, take the Automation practice questions. Aim for 80% or better before moving on to the full practice exams.