Multi-agent team coordination for large feature implementation

Implementing 8+ features across backend and frontend simultaneously is slow with sequential work. Shared files (models, main entrypoint, API client, type definitions) create merge conflicts when multiple developers work in parallel.

Group tasks by file ownership to minimize shared-file conflicts. One worker handles all model/schema fixes, another handles a feature that touches user models early, etc. For the remaining shared files (main.py router registration, models/__init__.py imports, API client), accept that multiple workers will touch them and verify consistency at the end. Write database migrations manually at the end after all model changes are complete rather than running autogenerate mid-flight. For Alembic migrations that add NOT NULL + unique columns to existing tables, first add as nullable, backfill data, then alter to NOT NULL and add unique constraint.

Parallel development is fastest when file ownership is clear. The bottleneck is always shared infrastructure files. By grouping tasks around file ownership and handling migrations last, you avoid mid-flight conflicts and ensure consistency.