Roadmap

Mako is an AI-powered SQL client today. Here’s where it’s going.

Current (V1)

What’s shipped and stable:

• Console — AI-powered SQL editor with schema-aware autocomplete
• Multi-database support — Connect PostgreSQL, MySQL, MongoDB, BigQuery, ClickHouse, Redshift, Cloud SQL, Cloudflare D1/KV
• AI Agent — Natural language to SQL, query explanation, schema discovery
• Query Runner — Execute, save, and share queries across your team
• Collaboration — Shared workspaces, saved queries, team access controls

Experimental

Features that work but are still evolving:

• SaaS Connectors — Pull data from Stripe, PostHog, Close CRM, and REST APIs into your data warehouse. Think of it as lightweight Fivetran built into your SQL client.
• Data Sync & Flows — Scheduled, resumable ETL pipelines with cursor-based pagination and automatic retries.

See the Experimental section in the docs for details.

Planned

What’s on the horizon (no timeline commitments):

• Dashboarding — Build visual dashboards from your saved queries. No need to export data to a separate BI tool.
• Reverse ETL — Write data back from your warehouse into SaaS tools (CRM updates, marketing lists, etc.).

Philosophy

The product grows along a natural complexity curve:

1. One database — Mako is already useful as a SQL client
2. Multiple databases — Connect all of them, query from one place
3. Data warehouse — Get your SaaS data in (connectors)
4. Write back — Push processed data out (reverse ETL)
5. Visualize — Dashboards on top of everything

Each step builds on the previous one. We ship each layer when it’s ready, not before.