Design schemas for change (MongoDB)

Steps

1. 1Identify entities and relationships; write 10 queries you must answer well
2. 2Normalize to 3NF by default; document any intentional denormalization
3. 3Use UUIDs or bigint PKs; never emails or natural keys
4. 4Add created_at/updated_at with DB-level defaults on every table
5. 5Constrain with FKs, CHECKs, and NOT NULL — let the DB defend integrity
6. 6Review growth estimates: which table hits 100M rows first, plan partitioning now
7. 7Model around access patterns: embed what reads together
8. 8Validate with JSON Schema at the driver level

Common Pitfalls

• ▲Soft-delete columns complicating every unique constraint
• ▲Storing money as float instead of integer minor units

Commands

$ npx skills add aniruddhaadak80/skills --skill databases-schema-design-basics-mongodb

$ npx skills add aniruddhaadak80/skills --skill databases-schema-design-basics-mongodb -g

Tags