Automatizav1
TEAM

People who take data seriously

A lean analytics-engineering squad, correctness-first, contracts by default, no vanity metrics.

TEAM · OPERATING MODEL

How the team works

Five engineers, one operating contract: every model has an owner, every dashboard has a metric tree, every alert routes to the table maintainer. We don't list people behind initials we can't share publicly — we list how the team operates.

  • OP-01

    Lean by default

    5 engineers across data engineering, analytics engineering, and BI. No middlemen, no offshore handoff, no SDR layer between you and the people writing dbt.

  • OP-02

    Operator-grade

    Median 8 years of warehouse production experience per engineer. Every project lead has shipped at least one dbt-first marts layer to a customer with on-call rotation.

  • OP-03

    Code review on every PR

    No model lands without an owner, schema tests, and a freshness contract. Pull requests get reviewed by a human who has seen the model break in production.

  • OP-04

    Postmortems published

    Internal. Not blog-bait. We write a five-line postmortem for every incident with a measurable corrective action, and we keep them indexed by table.

PRINCIPLES

Principles

P-01

Correctness-first

Before fast or pretty, it has to be right. Tests are contracts.

P-02

Source-of-truth-owned

One metric, one definition, one place. The semantic layer is non-negotiable.

P-03

Instrumentation-before-dashboards

Without instrumentation there is no dashboard, only guessing.

P-04

No-vanity-metrics

If a metric doesn't change a decision, it leaves the report.

WORKLOG · RECENT

Recent worklog

Git-log-style summary of deliveries across projects.

2026-04-18 14:02feat(dbt): semantic layer for NRR cohort on fintech client
2026-04-11 09:41chore(infra): move pipeline to hourly freshness + p95 SLA
2026-03-29 11:17fix(bi): replace broken funnel model in retention dashboard
2026-03-14 18:09feat(observability): contract tests for 12 critical sources
2026-02-28 10:23refactor(models): unified attribution model across 3 brands
2026-02-12 15:44docs(metrics): publish metric glossary v2 with owners
2026-01-27 09:06feat(alerting): anomaly detection on MRR deltas
2026-01-10 12:38chore(warehouse): migrate legacy cubes to dbt-first models

Wire a real pipeline to your warehouse

30-min stack review. We come back with a scoped plan, an owner, and a freshness contract. No SDR, no slide deck.

Book a 30-min reviewOpen the playground