Today, we're thrilled to announce the public beta of deet - a typed, versioned relational programming language designed to transform how you build, test, and deploy data pipelines.

The Problem We're Solving

If you've worked in data engineering, you've experienced the pain:

Runtime errors at 3am when a null value slips through your pipeline
Silent data corruption that goes unnoticed until a stakeholder reports wrong numbers
Deployment anxiety because there's no way to know if your changes will break production
Slow development cycles due to waiting for full DAG runs to test a simple change

Traditional SQL lacks the safety guarantees that modern software engineering takes for granted. dbt improved the workflow, but the underlying SQL remains dynamically typed and error-prone.

Enter deet

deet is built on a simple premise: your data pipelines deserve the same developer experience as your application code.

-- deet catches this at compile time, not at 3am
model users {
  select
    id,
    email,
    coalesce(name, 'Anonymous') as display_name -- Handles null safely
  from raw.users
}

model active_users {
  select
    u.id,
    u.email,
    u.display_name, -- Guaranteed non-null from above
    count(*) as event_count
  from users u
  join events e on e.user_id = u.id
  where e.created_at > current_date - interval '30 days'
  group by u.id, u.email, u.display_name
}

When you run deet check, we analyze your entire pipeline and catch:

Nullability errors before deployment
Type mismatches between joins
Missing columns in references
Circular dependencies in your DAG

Key Features

Compile-Time Type Safety

deet's type system tracks nullability throughout your entire pipeline. If you reference a column that could be null without handling it, you'll know immediately - not when your dashboard breaks.

Multi-Backend Compilation

Write once, deploy anywhere. deet compiles to:

DuckDB for local development and testing
PostgreSQL for transactional workloads
BigQuery for serverless analytics
Snowflake for enterprise data warehouses

Instant Deployments

Our cloud platform deploys your changes in seconds, not minutes. We use intelligent caching and incremental compilation to make iteration fast.

Git-Native Workflow

deet projects are just files in a Git repository. Use branches for features, pull requests for review, and merge to deploy. It's the workflow you already know.

What's Next

We're just getting started. Over the coming months, we'll be adding:

IDE extensions for VS Code and JetBrains with full autocomplete
Column-level lineage visualization
Data contracts for cross-team dependencies
Anomaly detection built into the platform

Try It Today

deet is available now in public beta. You can:

Try the playground at deet.dev/playground - no signup required
Read the docs at deet.dev/docs to understand the language
Join our Discord to connect with the community and get help

We can't wait to see what you build.

Have questions? Reach out on Twitter or join our Discord community.