Architect aims to make AI work from software plans, not raw code
Architect is an open-source project that tries to make the main instead of generated code. A high-level DSL is used to describe domain objects, services and dependencies, APIs, event consumers, deployments, , versions, and migrations. Architect parses that information into a fully linked model, which can later be used to generate code, tests, Docker files, settings, deployment manifests, and other project files.
The parser and model resolver are already implemented. The next planned work is the generation layer and a project lock file that records project structure, dependency versions, versions, generated artifacts, and task dependencies as a DAG. The intended workflow is AI-native: LLMs, currently tested locally with Qwen Coder, work on tasks and plans instead of directly editing source files.
The goal is more generation, so the system keeps knowledge about the codebase instead of asking an LLM to regenerate everything each time. A with embeddings would help find similar code for reuse and refactoring.
Key points
- Architect is an open-source tool for treating architecture as the project’s .
- The DSL describes services, dependencies, APIs, deployment, , versions, and migrations.
- The parser and model resolver already exist.
- The next pieces are and a lock file that tracks structure, versions, generated files, and task order.
- The AI workflow is meant to let LLMs act on tasks and plans rather than directly editing source files.