vectorient

How we work

From your systems to a dashboard you can trust, all written as code.

What happens to your information on the way from your point of sale, your ERP and your books to the dashboard your team opens in the morning, and who controls each step.

Your systems

  • POS
  • ERP
  • Accounting
  • E-commerce
  • Spreadsheets

Your cloud account

vectorient
  • Ingest
  • Model
  • Test
  • Control access

Your dashboards

  • QuickSight
  • Power BI
  • Looker Studio

In plain terms

Your company already has the information. It's spread out: sales in the point of sale, costs in accounting, stock in the ERP and the rest in a spreadsheet one person updates.

We build the layer that brings all of it into one place every day, in a cloud account under your name. That's where the totals get reconciled, where each metric gets its one definition and where you decide who sees what.

Your ERP and your other systems stay exactly as they are. We connect to them and read from them.

Today and with the platform

TopicTodayWith the platform
ReportsSomeone builds them by hand every weekThey refresh every morning
FiguresEach department has its own numberOne number, calculated the same way for everyone
HistoryLost when someone overwrites the spreadsheetKept, so you can compare year over year
AccessFiles travel by emailEach person sees what they should, and there's a record
ChangeA new store or system breaks the reportsIt joins the same model and the history stays intact

The architecture

Five layers. Each one does a job that's done by hand today.

  1. 01

    Ingestion

    Reads your point of sale, ERP, accounting, e-commerce and shared spreadsheets on a schedule. Nobody downloads a file again.

    Official APIs, database replicas or scheduled exports

  2. 02

    Storage in your cloud

    Every record in one place, with its full history, in an account under your company's name.

    Amazon S3, encrypted at rest

  3. 03

    Modeling and definitions

    Where returns, duplicates and timing get resolved, and where margin, fill rate and days of inventory are defined once.

    dbt, with tests on every model

  4. 04

    Automation and checks

    Runs the whole process every day, checks that the totals tie to your books and alerts us when something doesn't.

    GitHub Actions, data quality tests, alerts

  5. 05

    Dashboards by area

    Sales, inventory, purchasing, suppliers and finance, each built around the decisions that team makes.

    QuickSight, Power BI or Looker Studio

Everything is written as code

Every piece of the platform lives in a repository your company owns.

  1. 1

    The infrastructure

    Accounts, storage, permissions and networks, declared in Terraform. Any environment can be rebuilt from the repository.

  2. 2

    The business logic

    Every metric is a dbt model with its definition and its tests, readable by anyone on your team.

  3. 3

    Every change

    Goes through a pull request, gets reviewed and deploys automatically. The history shows who changed what and why.

The code is written with AI engineering tools. Every definition and every number is reviewed and signed off by an architect before it reaches your dashboards.

The tools we use

Proven, widely used tools that your team or any future hire already knows.

AWS
The cloud where the platform runs, in an account under your name.
dbt
The business logic: how each metric is calculated, versioned and tested.
Terraform
The infrastructure, written as code and reproducible.
GitHub Actions
Runs the daily process, the tests and every deployment.
QuickSight, Power BI or Looker Studio
The dashboards. We use the one your team already works in, or pick one with you.
Databricks and Unity Catalog
For companies on Databricks: the platform and its governance.

Already on Databricks, or ready to move

A dedicated line for companies that run on Databricks or have outgrown a simpler setup.

  • Platform

    Workspaces, networking and compute on AWS, declared in Terraform and sized to your volume.

  • Governance

    Unity Catalog with access by team, by area and down to a single column, with a full audit trail.

  • What comes next

    Genie spaces where your teams ask the data in plain English, and forecasting models for demand and stock.

We've built Databricks platforms for several business units under one governance model, with access controlled down to the column.

Security and access

You'll be giving us access to sensitive information. Here's what happens to it.

  • We read your systems

    We connect to your point of sale, ERP and other sources with read-only access, through their official connection, a database replica or the files they already export.

  • We work through access you control

    Dedicated users inside your account, with only the permissions they need. You see what they do and can revoke them at any time.

  • Everything is encrypted

    In transit and at rest. System credentials live in a secrets manager in your account.

  • Everything is logged

    Every query is recorded: who saw what and when, on your team and on ours.

What's yours

  • The cloud account

    Under your company's name, billed to you directly by the provider.

  • The repository

    All the code, infrastructure and definitions, in your company's GitHub.

  • The documentation

    How it's built, how it runs and what each metric means, ready for your own team or whoever you choose.

Common questions

Do we have to change our ERP or our systems?

No. The platform reads from what you already use. Your operations stay the same.

Where does our information live?

In a cloud account under your company's name. We build and run the platform there, and it's yours.

Do we need to hire a data team?

Not to start. We run the platform and keep it right. If you build your own team later, they take it over with the code and the documentation.

We already use Power BI. Is this the same thing?

Power BI is where you see the numbers; the platform is what makes them correct and keeps them current. You keep using Power BI on top of it.

What does "kept right" mean after the first dashboards?

Your business changes: a new store, a new system, a new question from the board. We add the source, update the definitions and keep the checks that tie every total to your books.

Do we need Databricks?

Most growing companies run well on AWS alone. Databricks makes sense if you already use it, handle large volumes or have several teams working on the same data. We decide it with you in the diagnostic.

Can we use AI on our data?

Once the data is reconciled and defined, yes: you can ask it questions in plain English and get answers you can trust. We build the foundation first so those answers are right.

Let's find the number you need and can't get.

The first conversation is about which systems you use, which reports your team builds by hand and which question goes unanswered.