Recently I had the honor to discuss AI, testing, and the DORA AI Capabilities Model with the leading practitioners who host the Beyond Quality podcast. Hosts Maryia Tuleika, Vitaly Shapovalov, and Anupam Krishnamurthy are all experienced in using GenAI and LLM technology. I took away so many insights. For example, how agents degrade over time (so we need to continually test our agents, as in this video from Antony Marcano). And the many dangerous security pitfalls, which do not seem to be on everyone’s radar. As a “guide” for the DORA community, I especially appreciated our conversation about how the AI capabilities model can guide your GenAI journey. And best of all – these three, expert in using AI and actively researching how to benefit most from it, confirm my hope that pairing and ensembling is easier and more needed than ever.
Please watch on YouTube or listen on Apple podcasts or your favorite podcast provider. And, I highly recommend reading the ongoing results of Beyond Quality’s research, such as this on QA in the age of AI-accelerated development.
If any of you readers is willing to share some of what you are doing with AI technology, I would love to learn from you. I currently don’t have the opportunity to work on a team using agentic QE. I think it can give us an amazing opportunity to increase collaboration across different specialties in software organizations and I’d love to see if that’s happeningn.
