I recently started an AI learning journey, experimenting with various AI tools to understand their capabilities and limitations. My focus has been on grasping core AI concepts like large language models, neural networks, and prompt engineering to better understand how generative AI works and how to use it responsibly.
I was inspired by Henrik Kniberg’s March 28th article entitled “Agile in the Agile of AI” about using AI as a virtual team member. In the article, Henrik explores what the future of Agile teams would look like using AI as a team member. I encourage you to read the article since he shares an interesting perspective on the potential impact of AI to Agile Teams. I also recommend watching his Generative AI in a Nutshell video on YouTube since it provides an excellent overview of the topic: Generative AI in a Nutshell - how to survive and thrive in the age of AI - YouTube.
I recently spent a few hours one afternoon exploring what it would be like to have an AI agent as a member of an Agile Team. I decided to keep it very high level and fun. My wife and I have several cats who love to chase after thrown toys, so I started with a very basic premise by creating a fictitious team called Team Furball that is working on an automated cat toy launcher product called Katapult.
Team Furball decides to experiment with using an AI agent to help create a feature for their product using the Scaled Agile feature template. They worked with the AI agent to break the feature into user stories with acceptance criteria, generate acceptance tests from the acceptance criteria using the Gherkin language, and then generate the XML and Java code to run the tests in the Cucumber framework. I was really impressed with how well the AI agent performed, providing accurate and detailed responses that Team Furball could use, refine, and build upon to accelerate the implementation of their Katapult product.
Taking the experiment a step further, I used a video-generating tool to create a video to demonstrate Team Furball using AI as a virtual team member. The video was made using the free version of Synthesia. Synthesia uses an AI avatar to narrate the video. The AI agent in the video is ChatGPT, and the images were created using DALL-E 3.5 within Copilot.
This exercise exposed me to a small set of effective AI tools. Please keep in mind this is a first attempt, so it’s a bit rough, but it was fun to create. Check it out and let me know your thoughts: Click here to watch the video.
I encourage you to learn more about AI and how it could assist you and your team in your day-to-day work. The more you learn about the technology and what it can and cannot do effectively at this stage helps to demystify it and opens up a lot of possibilities to be more creative and product. Enjoy!
Tom Wessel brings over 30 years of experience in various industries in software development. His experience spans the entire end-to-end product life cycle from inception to delivery. With over 15 years of experience as a trainer and coach, Tom has worked with various organizations of all shapes and sizes to train, coach, and mature their agile mindset and transform how they deliver value.
View Tom's calendar here to schedule an introductory call.