Context & EvalsKeep your models, metrics, and skills accurate
We draft your dbt models, semantic models, and skills from your dbt project, BI tools, and agent traces. Issues surface from every chat, an agent drafts the fix, and you approve it once evals pass.
Drafted from your stack
We ingest your BI tools, dashboards, dbt project, and agent traces, and draft the context your agents need, as code. Your data team curates the draft.
- dbt models
- Semantic models, Apache Ossie compatible
- Skills
- Evals
Issues surfaced as your team chats
While the agent answers questions, it flags gaps and inconsistencies and clusters them into suggestions, with hits, people, and votes. An agent drafts each fix for your review.
No churn definition yet, so I drafted one for your data team.
Every change gated by evals
Each fix is a pull request that links the chats, votes, and evals behind it. Jev, our type-safe model, checks every revision, only the evals the change touches rerun, and an admin merges when they pass.
Gold values are reconciled against your warehouse, never against the agent's own answer.
feat: refund reason on int_orders_with_refunds #212
signalpilotproposed · feat/refund-reason → main
- add refund_reason to int_orders_with_refunds
- fix: keep 1:1 grain on order_id · signalpilot
- ✓Dana R. approved these changes
Bring your dbt project. We will show you what it already knows.
Open source · Apache 2.0 · GitHub · Slack