Hands-On with the AI-DLC Unicorn Gym Workshop - Learning AI-DLC Through Team Development
A complete look at the hands-on workshop where you experience AI-DLC through three days of team development. Learn how to run Mob Elaboration and Mob Construction, and how to leverage open-source workflows.
What Is AI-DLC Unicorn Gym?
AI-DLC Unicorn Gym is a hands-on workshop led by AWS for practicing the AI-Driven Development Lifecycle (AI-DLC). Participants form cross-functional teams and apply the AI-DLC methodology to real projects over three days. Rather than learning AI-DLC theory in a classroom setting, participants write actual code using Kiro and Amazon Q Developer while experiencing the three phases of Inception, Construction, and Operation firsthand. The workshop is designed for all experience levels, from seasoned engineers to team members who don't typically write code.
The Three-Day Schedule
The workshop spans three days. Day one begins with an introduction to AI-DLC core concepts and the tools used, followed by team formation and project assignment. Each team enters the Inception phase, feeding business requirements into AI and using Mob Elaboration to validate and refine the AI-proposed requirements and user stories as a group. Day two moves into the Construction phase, where teams evaluate and approve AI-proposed architecture and code through Mob Construction while progressing the implementation. Participants also practice configuring steering files and Project Rules to align AI output quality with organizational standards. Day three covers the Operation phase and deployment, concluding with each team presenting their results.
Leveraging Open-Source Workflows
The AI-DLC Unicorn Gym leverages the open-source AI-DLC workflows published by AWS (awslabs/aidlc-workflows). This repository contains Amazon Q Developer Rules and Kiro steering files that implement the three principles of AI-DLC: adaptive workflow selection, flexible depth adjustment, and built-in human oversight. Simply conveying your project intent in natural language allows AI to evaluate the context and automatically select the appropriate phase and depth. You can experience adaptive behavior such as applying a lightweight workflow for simple bug fixes and a detailed workflow with full requirements and design for new feature development.
Rolling It Out at Your Organization
The insights gained from the workshop can be directly applied to introducing AI-DLC to your own development teams. Bring back steering file templates, Mob Elaboration and Mob Construction facilitation procedures, and team formation best practices to apply to your own projects. As an adoption benefit, traditional sprints (weeks) are compressed into Bolts (hours to days), accelerating development velocity. At the same time, quality and governance are maintained because human approval is built into every decision point. Consult your AWS account team to customize the workshop to your organization's specific requirements.
How to Think About the Three Phases
AI-DLC views development in terms of three phases: Inception, Construction, and Operation. In Inception, starting from business requirements, you clarify what should be built through dialogue with AI. In Construction, AI proposes the design and code, and the team advances implementation while evaluating and correcting them. In Operation, you carry out deployment and operations, and look back on the insights gained. What matters is not making every phase uniformly heavy, but adjusting the depth according to the nature of the problem. By varying the intensity, keeping simple fixes light and handling complex features carefully, you achieve both speed and quality.
The Value of Working in Mob Style
In the workshop, you practice Mob Elaboration, in which the whole team verifies the requirements definition, and Mob Construction, in which the team advances implementation while evaluating it. By reviewing the deliverables that AI proposes from multiple perspectives instead of letting individuals keep them to themselves, you can discover oversights and misunderstandings early. When less experienced members also join the discussion, tacit knowledge is shared and the understanding of the whole team deepens. This way of proceeding, in which the team examines AI output collectively instead of accepting it at face value, achieves quality assurance and knowledge transfer at the same time. Through collaboration, the team can acquire a feel for making full use of AI.
The Importance of Building In Human Approval
At the core of AI-DLC is the principle of inserting human approval into every important decision. AI performs proposals and implementation at high speed, but it is the role of humans to judge whether the content matches the requirements and meets quality standards and conventions. If you convey your organization's standards to AI in advance through steering files and rules, you can align the direction of its output. On that basis, by having people review and approve at each stage, you can maintain governance and quality while taking advantage of speed. In the workshop, you can learn hands-on how to balance automation and control.
Preparing to Participate and Forming Teams
Preparing in advance helps make the workshop fruitful. If you set up the environment for the tools you will use and grasp the outline of the project you will work on, you can get started smoothly from day one. If you make the team cross-functional by including not only engineers but also members who know the business and the product well, discussions of requirements definition become richer. A feature of AI-DLC is that people who do not usually write code can also take part in development through dialogue with AI. By bringing together diverse perspectives, you can gain results grounded in real problems and concrete lessons for rolling it out at your organization.
Summary
AI-DLC Unicorn Gym is a three-day team development workshop that provides hands-on experience with Amazon's AI-DLC (AI-assisted Development Lifecycle). Participants learn development processes that leverage open-source workflows and take home concrete techniques and team management know-how for integrating AI into their own product development.
References (Official AWS Resources)
The primary sources for this page are the official AWS website and documentation. Check the official pages below for the latest specifications and pricing.
If this page and the official documentation disagree, treat the official documentation as authoritative.