Cisco Live 2026 Through a DevOps Lens

I was fortunate to attend this year’s Cisco Live conference in Amsterdam, and as a DevOps consultant at ITSUS working on a range of network automation and secure‑infrastructure projects, I was looking forward to seeing the latest developments in Cisco’s automation and AI tooling. Cisco Live provided the ideal opportunity not only to update my certifications, but to gain firsthand insight into how automation and AI are becoming inseparable from modern network engineering.

Below, I’ve captured my key takeaways from the event, especially those relevant to defence and secure government environments where ITSUS delivers much of its work.

Automation and AI in the Spotlight

This year, Cisco Live brought closer together the newer DevNet (now Automation) and traditional networking technologies, with the conference focusing on the benefits of AI and Automation to network engineers. It’s probably no coincidence that the timing of the conference aligned with Cisco’s refresh of their learning materials and exams, as Automation (and AI) is now woven throughout each Cisco offering in various ways – with the key takeaway that (Net)DevOps is no longer a separate arena to traditional networking; (Net)DevOps technologies and methodologies are integral to robustly control and maintain large networks.

Following on from last year’s Cisco Live 2025, AI and Automation were front and centre through the conference. The majority of sessions made reference to areas AI and specifically AI with Automation could be used as a complementary tool for network engineers, whilst showing due care and diligence on the part of the Cisco engineers delivering the sessions, who are well aware of the potential caveats and pitfalls – such as the need for verification and validation of AI-based suggestions and how AI could be destructive to parts of the network if not used carefully.

Cisco and AI-Supported Automation

The use of AI-supported automation to provide greater network insights (e.g., through network monitoring) and support for network programmability stood out to me as the main approach Cisco is taking. Not only does this use of AI reduce the barriers to entry for non-programmer network engineers, but it helps provide additional checks and information for network operations.

Key Platform Updates to Note

One thing that stood out for me was that Cisco were giving more attention to their Cisco Modeling Labs (CML) platform. Having used this platform already for a number of years, it was great to see some more of the changes put in to make it more accessible and user-friendly, as well as making it a tool that could be used to supplement documentation alongside its (arguably) most useful function, that of acting like a digital twin.

With updates to the CML API, automation and AI-based tools can potentially leverage CML to run automated testing and deploy novel network configurations for engineers to explore, evaluate and learn from.

Automation in Air‑Gapped and Secure Government Environments

One of the most pleasant surprises of the conference was in attending multiple sessions that talked about AI and Automation and mentioned their usage in air-gapped environments. For defence and secure government networks, air-gapped environments are the norm, so it’s key for any adoption of new technologies that they can function – and function well – without Internet connectivity.

People might assume that AI is only relevant when leveraging the cloud-based products of well-known large-language models (LLMs) or similar, so it was great to hear of how the benefits of these complementary tools could be applicable to a wider range of networks. Automation of network insights and testing are two areas that are incredibly important for secure government networks – especially those that form critical national infrastructure, where the network needs to be resilient and stable, and operators have access to all the information they need to ensure it remains this way.

The Rise of Infrastructure‑as‑Code and Centralised Configuration

Consistency in defence networks is one of the most important aspects of network design, and there is definitely a push to adopt the benefits of Infrastructure-as-Code (IaC) and centralised management and deployment of configuration.

Products that assist in facilitating this – and which work in air-gapped environments, such as Cisco Catalyst Center – are helping drive adoption of IaC and automation. Configuration stored in Catalyst Center can be automatically deployed to Cisco network devices within the fabric of the network and can monitor/record changes that have been made to configuration both within and outside of this centralised source of truth (e.g., by an individual at the device level). Tools such as Catalyst Center can help provide audit trails and ensure consistency of devices by alerting operators to deviations in configuration and as a result potential compliance and inconsistency issues.

Skills, Barriers, and the Evolving Network Engineer

Cisco automation requires engineers with skills in both traditional networking and in areas such as programmability and working with APIs. This means that organisations truly benefit when they have engineers who can do both – or a mix of engineers from each background who work well as a team and have at least a basic understanding of each of the technology groupings.

As seen at Cisco Live and across the board with the directions Cisco is taking, the line between traditional networking and automation is reducing more and more, and automation is becoming an important pillar in Cisco’s products – so having multi-disciplinary network engineers or network teams is becoming a requirement rather than an optional extra.

Preparing for the Next 2–3 Years

AI is the big disruptor of the moment, and organisations will need to know how to use it well where appropriate – and also where it is not appropriate to be used, or not sufficient by itself. The challenge for organisations is twofold – to understand the benefits and the risks of AI and to ensure that they know how to use it where it’s applicable in enhancing a solution and where not to use it.

At Cisco Live, Cisco’s engineers showed that they are aware of both positives and potential negatives for AI usage, and treated it as a tool that supplements, rather than replaces.

Where We Are Focusing Next?

Whilst DevOps is a topic in its own right, with standalone solution options that we can and have delivered for customers, I think we need to highlight the importance of cross-discipline solutions such as NetDevOps and demonstrate our capability in providing teams with engineers from multiple diverse backgrounds. Diversity in teams often increases the pool of knowledge and experience and helps us provide a stronger offering to customers. As technology areas merge or link, having this broader knowledge base will strengthen our team and strengthen the service we can deliver to our customers.

ITSUS has been ahead of the curve in building its team of consultants to cross multiple subject areas and has worked on a number of projects in recent years with both network and automation challenges. We have a strong multi-disciplinary team with a wide range of experience and expertise and have repeatedly delivered or helped deliver automation and network solutions to a variety of customers, at a high standard, within budgets and on time.

Conclusion

Cisco Live 2026 has made it clear that automation and AI are no longer niche specialisms, but core components of modern network engineering. For defence, government, and secure environments, Cisco’s investment in air‑gapped automation workflows, enhanced modelling tools like CML, and platforms such as Catalyst Center highlights a future where networks can be more resilient, predictable, and consistent without sacrificing security.

With multidisciplinary expertise across both networking and automation, ITSUS is definitely well positioned to help organisations adopt these advancements safely, effectively, and with confidence.