In the March 2025 edition of AWS Innovate, Mark Relph, an AWS executive, presented "7 principles for effective and cost-efficient generative AI apps." The presentation was based on recurring themes in his discussions with AWS customers. For Data and AI leaders within IT organizations, these principles offer valuable guidance in setting a strategic direction. Whether you are learning to navigate the Gen AI landscape or under pressure to demonstrate the value of Gen AI projects, these principles are relevant, irrespective of your cloud service provider. A quote I need to highlight: "Don’t underestimate the work to have your data Gen AI ready. Pick tools to make this easier." Data preparation is 80% of the effort, and AI implementation 20%.

Link to the presentation.

Scaling AI requires a whole different level of effort than prototyping it. If you are stuck in proof-of-concept purgatory, it is time to stop looking for perfection. Take steps towards deployment by having clear goals, building an AI Ops team with internal resources, securing your data and intellectual property, providing safeguards, and educating the business.   

Link to the presentation.