Webinar: AIOps and MLOps Applied to Platform Engineering

AI as Your Ally

Platform engineers, AI doesn’t have to be the enemy. In fact, it can be a powerful ally, transforming your role and streamlining your day-to-day operations. To stay ahead of the curve, it’s crucial to understand how AIOps and MLOps can transform your role.

In our recent webinar, John Boero, Field CTO of TeraSky UK, demystified AI for platform engineers and provided actionable insights into how to leverage these technologies effectively.

The webinar focused on the capabilities and limitations of large language models (LLMs) in the context of software development. John discussed prompt engineering, which involves crafting effective prompts to control the behavior of LLMs. He emphasized the importance of understanding the differences between completion and chat models, as well as the ability to customize prompts for specific tasks.

John also covered the use of LLMs for code generation and analysis, demonstrating how LLMs can be used to write code snippets, translate between programming languages, and even provide explanations for code segments. However, he also highlighted the limitations of LLMs, such as their tendency to “hallucinate” or produce incorrect code.

The webinar then turned to the potential of LLMs to automate tasks related to code building and deployment. Building and deploying code can be a manual process involving multiple steps like selecting requirements, checking documentation, building templates, and testing. New LLM models with integrated “rag agents” can automate these steps. The model can interact with web services to retrieve information (e.g., documentation) and integrate it into the code generation process. In this way, automation can significantly reduce manual work and improve efficiency. It can also potentially improve accuracy by reducing human error. 

John demonstrated this potential by comparing two LLM models – Bedrock, an AWS service which offers a user-friendly interface and seamless integration with other AWS tools and services, and Gemma, developed by Google DeemMind, which offers more options for private deployment in the case of strict data privacy requirements. 

Ready to unlock the power of AI for your platform engineering team? By leveraging the insights and strategies discussed in our webinar, you can streamline your workflows, improve efficiency, and keep your edge. Watch the recording now.