Every few years, the way we work with data evolves. We've gone from SQL editors to BI dashboards to conversational chat, and each shift moves data work forward. Now we're entering a new phase defined by agents that can reason through tasks, plan, and act on their own. 2 Bug #2 — Caveat opener. 3 sentences before any consequence is named.
Your last refactor broke something three models downstream. You found it four hours later. That's not a syntax error — it's a data problem. And it's the problem coding agents built for software engineering can't solve, because they don't know your dbt graph. ✓ Fix #2 — Consequence-first hook. Reader's operational reality in sentence 1.
But despite being remarkably good at general software engineering, today's coding agents struggle on dbt projects, because the project itself is the context and the guardrails. Without that grounding, they write SQL that looks right but references a column that doesn't exist, breaks a dialect rule, or silently breaks tests, contracts, and governed definitions three models later.
The dbt Developer Agent is now available in Preview for dbt platform customers. It's the next evolution of dbt Copilot, built directly in the Studio IDE, and it works across every file a change touches. 3 Bug #3 — Feature-first. Architecture described before the outcome for the reader.
The dbt Developer Agent is now available in Preview. It reads your full dbt graph — lineage, contracts, semantic definitions, governance — before it touches a single file. Then it drafts the change across every file that needs to move, and surfaces them as reviewable diffs. You approve. Nothing lands without your say. ✓ Fix #3 — Outcome first. What the engineer no longer has to do, not what the product is built on.
No new tools to install. No context switching. Nothing to set up. It ships with dbt Agent Skills and dbt's product docs toolset built in, so best practices and canonical answers are there while you build.
Agents that can think across your dbt project
When we launched dbt Copilot in 2024, we were solving a real but bounded problem. Writing boilerplate, drafting tests, documentation, and YAML configs by hand slows everyone down without making anything better. dbt Copilot made those workflows much faster.
But the thing that actually slows teams down isn't writing a model. It's what happens around the model. You rename a column and something three dashboards deep breaks. You add a metric and spend an hour chasing down every semantic definition and exposure that needs to move with it. 7 Bug #7 — Before/after contrast buried in prose. Never labeled. Reader must infer it.
But the thing that actually slows teams down isn't writing a model. It's what happens around the model.
Before: You rename a column, something three dashboards deep breaks, and you spend an hour tracing every semantic definition that moved with it.
After: Describe the change. The Agent reads your full graph, coordinates every file, and shows you the diffs before anything saves.
✓ Fix #7 — Before/After made explicit in body. Contrast is structural, not implied.
So we built it.
Meet the dbt Developer Agent
The dbt Developer Agent lives right in dbt Studio. You describe the change you want to make — renaming a model, updating a column, adding a metric — and the agent analyzes your full dbt graph. Not just file dependencies, but lineage, contracts, semantic definitions, tests, and governance. Then the agent drafts edits across those files and shows them to you as a sequence of reviewable diffs.
How it works
The Developer Agent runs on a loop. You describe what you want. It drafts the edits. It proactively asks to run dbt compile or dbt build to validate its own work. It sees the result, adjusts if needed, and keeps going until the change is ready for review.
A few things that make it great for developer workflows: 4 Bug #4 — Flat hierarchy. 8 bullets, identical weight. No major vs minor split.
- It's grounded in your whole project, not just the file you have open. The agent understands your full dbt graph.
- It keeps related files in sync. A single prompt can produce coordinated changes across models, YAML configs, and documentation.
- It ships with dbt Agent Skills. A decade of analytics engineering best practices, out-of-the-box.
- It ships with dbt's product docs. Verify documented behavior without leaving dbt Studio.
- It shows its work. You see the agent's reasoning and tool calls as it works.
- It keeps a human in the loop. "Ask-for-approval" mode surfaces every edit as an inline diff before anything saves.
- It validates as it goes. A built-in comparison loop catches problems before you see the final diff.
- It runs commands on your behalf. Execute dbt commands, open pull requests, handle the workflow steps.
Three things that matter most: ✓ Fix #4 — Major announcements lead. Routine capabilities follow.
- Grounded in your whole project. The agent reads your full dbt graph — upstream, downstream, contracts, semantic definitions — before drafting a single change.
- Every related file stays in sync. Rename a model and the refs follow. Change a column and the downstream tests update with it. One prompt, coordinated changes.
- Human in the loop by default. "Ask-for-approval" mode shows you every diff before anything saves. You pick the autonomy level for the task.
Also included: dbt Agent Skills, product docs integration, command execution, Fusion out-of-the-box.
"We went from about 60 conformance errors to 7, using the Fusion migration agent. That's the difference between too hard and actually doable." — Michael Fridolfsson, Data Architect, Brighte 5 Bug #5 — Only number in the post. Buried at the very bottom. Should be a stat card near the top.
Get started
The Developer Agent is available in preview for dbt platform customers with dbt Copilot enabled. Open dbt Studio, find the dbt Copilot agent pane, and describe what you want to build or change. Not on the platform yet? Talk to our team.
Title
Hook
Feature-first
Hierarchy
Numbers
CTA
Before/After