Every vibe coding tool out there starts building the code immediately as they get some sense of the product. But product development is not just code generation. There's so much that the Ai needs to understand about the user, market, goal of the product before developing it. I bring my workflows to build product-market intelligence in Meridian which helps users solve real problems by automating research, scoping ideas and building products.
Would love to hear all the different problems you are facing in spec driven development. When feature priorities change or ad-hoc bugs surface, do you create new specs or modify old ones ? How is the kanban board tracked during spec driven development ? I feel the kanban board statuses are not suitable for spec driven development and we need to define a custom workflow to make it work.
Meridian helps AI-native teams turn customer research into specs their human and AI builders can execute.
The platform helps to discover new feature ideas by running surveys on Linkedin; prioritizes and scopes features into tracks; prepares a plan where humans and Ai agents can work collaboratively to complete the track; release it and monitor for feedback to close the loop.
Create a project with Astra and it will help you to know your product before you build it.
