• You and AI News
  • Posts
  • Converging network automation with Artificial Intelligence

Converging network automation with Artificial Intelligence

A fifth age of networking: the artificial intelligence age

My name is John Capobianco, and I serve as the Developer Advocate for Cisco Learning and Certifications Training Bootcamps. While my official role keeps me busy, my passion for AI and development doesn't stop there. Over the past year, in my free time, I have delved deep into the world of AI, specifically focusing on integrating OpenAI's technologies into Cisco's network automation platforms.

I started by incorporating OpenAI's ChatGPT APIs into Cisco's pyATS jobs. pyATS, being a robust framework for network testing and automation, offered a unique opportunity to enhance its capabilities with AI. Integrating ChatGPT APIs was a challenging yet rewarding endeavor, allowing us to automate and streamline network testing processes with a level of intelligence and efficiency previously unattainable.

Moving forward, I developed working ChatGPT plugins for Cisco platforms. This advancement was a game-changer, enabling Cisco's network systems to interact and respond to natural language queries. These plugins marked a significant step towards making network management more intuitive and user-friendly.

My most recent achievement has been developing working GPT models for various Cisco platforms and network automation tasks. These models, equipped with conversational code builders, have been a leap forward. They not only understand and generate human-like text but also allow for the incorporation of backend APIs through Actions using an openapi.yaml specification. This approach mirrors the one used in plugins, making it a cohesive and integrable solution across Cisco's systems.

Building these AI tools, especially the GPT models, has been a complex process, but the results are immensely gratifying. The ease of integrating backend APIs, coupled with the conversational nature of the code builders, has opened up new avenues in network automation.

In conclusion, with the advent of easily accessible AI in the form of APIs and Python SDKs, we are undoubtedly entering a new era in networking - the fifth age, the Artificial Intelligence Age. This era promises unprecedented advancements in how we interact with and manage network systems, driven by the power of AI and automation. As we step into this new age, the possibilities seem limitless, and I am excited to be at the forefront of this transformative journey.