Data consultancy
I build the pipelines, models and reporting underneath them, so your team spends its meetings arguing about decisions instead of definitions.
Four things, usually in this order. Most engagements start because a report stopped being believed.
Raw landing tables, incremental MERGE loads, keypair auth, and roles and warehouses that don’t surprise you on the bill. The unglamorous layer everything else sits on.
Sources to marts to reporting. Snapshots for the facts that change quietly, tests that fail loudly, and metric definitions written down where the whole team can read them.
The APIs no off-the-shelf connector covers — Shopify Admin GraphQL, HubSpot, Salesforce, ChartMogul, Google Ads — with resumable backfills for the history you actually need.
Daily operational numbers, weekly trading reviews, a semantic layer when it earns its place. Delivered where your team already looks, not in another tool to remember.
Small engagements, visible increments, and an exit that doesn’t leave you stranded.
About a week reading your stack and talking to the people who use the numbers. You get a written account of what’s broken and what it would take to fix — useful on its own, even if we stop there.
Agreed deliverables, shipped in increments you can see. Everything runs in your warehouse and your CI from the first week, never on my machine.
Runbooks, documented metrics, and a repository your team owns outright. The goal is that you don’t need me on retainer.
DTC retail · Shopify → Snowflake
Shopify Admin GraphQL extracted into Snowflake with MERGE upserts, dbt marts and snapshots layered on top, and a resumable backfill covering years of fulfilment and tracking history. Daily inventory metrics and weekly brand-and-category trading reports now land on their own. The whole pipeline runs on scheduled GitHub Actions, so there is no box to maintain.
Usual tools: Snowflake · dbt · Python · GitHub Actions · Cube · DuckDB
A short email is enough: what you’re running, the question you can’t answer, and when you need to answer it.
hello@sidewalkai.app