Database design. A data model that stays fast and correct as you grow.
We design or repair the data model behind your product: schema, indexes, access rules, migrations and backups.
For products with slow queries, inconsistent data or a schema that grew without a plan.
Database design
We review how your data is used, design a schema that reflects it, add the right constraints and indexes, set up row-level access rules where needed and plan safe migrations. Backups and restores are tested, not assumed.
- (01)A schema that matches the domain
- (02)Indexes and query tuning
- (03)Access rules and auditability
- (04)Safe, reversible migrations
Tools we use for this.
- PostgreSQL
- Supabase
- Row-level security
- pgvector
- Prisma
- SQL
We choose the stack that fits your requirements. Tool names describe our skills, not partnerships.
Setup.
1–2 weeks.
One clearly bounded piece of work, such as a system configured or a review completed, handed over ready to use.
Included
- (01)Data and query review
- (02)Schema design and documentation
- (03)Index and performance tuning
- (04)Access-control rules
- (05)Migration plan and scripts
- (06)Backup and restore test
Not included
- (01)Ongoing database administration
- (02)Licence and hosting costs
- (03)Rewriting application code beyond the data layer, unless scoped
How it runs
- (01) · Days 1–3
Review
Queries, data quality and growth patterns.
- (02) · Week 1
Design
Schema, rules and migration plan.
- (03) · Week 2
Migrate
A tested migration with a rollback path.
Also available as: Sprint (4 weeks). Compare the formats.
Can you fix our existing database without downtime?
Usually, by migrating in small, reversible steps. We plan the order and the rollback before touching production.
Do you work with databases other than PostgreSQL?
Our default is PostgreSQL. We can review and work with others, and we will say plainly when a migration is worth it.
Do you handle vector search for AI?
Yes, for example pgvector in PostgreSQL, when your AI features need search over your own content.