Synopsis:

Engineering delivery can feel like chaos that is unpredictably orchestrated. In high-growth companies, it gets harder and harder to answer the age-old question, when can we expect to ship our next release? We explore whether AI can help us predictably and reliably get to dynamic delivery; a GPS-like system that is constantly adapting to changing teams, customer requirements and finding the best route for delivery.

Talk abstract:

As companies grow, delivering software becomes increasingly challenging, and answering the question "when can we expect to ship our next release?" seems more like guesswork than science.

However, AI can help us navigate complexities by providing a GPS-like system that's constantly adapting to changing teams, customer requirements, and finding the best route for delivery. This talk will delve into the potential benefits of using AI for engineering delivery, including better predictability, benchmarking in teams, and the ability to learn from past successes and failures in agile.

We'll explore whether a metrics-based approach can help AI-led systems understand optimal routes for delivery, with an emphasis on DORA metrics such as deployment frequency. We'll also discuss the challenges of adopting AI in engineering delivery, including data quality from issues on Jira or pull requests from Github, and the cultural shift required for successful adoption.

By the end of this talk, you'll have a better understanding of whether AI can help with predictable and reliable engineering velocity. You'll leave with actionable insights on delivering value to customers, and an innate curiosity around the application of AI to product and engineering delivery.