Easycoder, working with AWS, ran a two-day hands-on workshop with a New Zealand property management and tenancy platform. The workshop applied the AI-Driven Development Lifecycle (AI-DLC) to priorities already on the team's roadmap.
Most engineering teams have tried AI for writing code. The harder part is using it to make good product decisions: what to build, how it fits with existing systems, and what it will take to run. This engagement tackled that question on real work, so the team could build skills they can keep using.
A structured approach to AI-driven development
AI-DLC treats AI as a working partner across the whole delivery process. AI breaks down goals, asks clarifying questions, proposes designs and generates code. The team reviews each output, weighs the trade-offs and makes the important calls.
The method runs across three connected phases:
• Inception: turning business intent into requirements, acceptance criteria and units of work the team reviews together.
• Construction: designing, building and testing those units, with engineers checking the results.
• Operations: supporting deployment, monitoring and maintenance, with AI recommendations and people in control.
Requirements, plans and decisions carry context from one phase to the next. Work moves in short cycles with frequent checks, so the plan stays grounded in how the real system behaves.
Applying the method to real customer priorities
The team worked on two priorities from their roadmap, inside a platform they already knew well. Over two days they moved from defining the problem to planning and building practical solutions, with the needs of property managers kept at the centre of every discussion.

A key moment came on day two, when the team brought in fresh thinking from the wider business and took a second look at their plan. Because the requirements and decisions were already written down through AI-DLC, the team could test each option quickly against what their integrations could support, what it would cost to run, and what users would experience. They chose a clear path for each priority and split into two streams to keep both moving.
The structured format also surfaced questions about data, billing and user flow early. These are the kinds of questions that usually come up mid-build, when changes are slow and expensive.
Practical experience to build on
By the end of day two, both streams were working productively with AI-DLC, and the team planned to keep using it in future projects. The workshop showed that clear problem statements, good documentation and shared context help AI give better results.
It also showed the value of getting product, design and development in the same room from the start. Testing ideas against real constraints before anything is designed leads to faster decisions and fewer surprises later.
Easycoder runs every AI-DLC workshop on the customer's own priorities, so teams leave with real progress on their roadmap and a method they can keep using.
Have a development challenge or AI project in mind? Get in touch with the Easycoder team.



