AI Automation Data Software

Writing code that does your job.

BrandUp Factory is a technology company from Hamburg. We build AI agents and custom AI, automate business processes, engineer the data underneath and ship our own software products.

Agent run

Task

Go through today's return requests and prepare the replies.

Runs locally · customer data stays in your environment

Plan

  1. Read the open return requests
  2. Look up each order and its stock
  3. Check the return policy
  4. Draft a reply for each customer

Activity

Working

  • shop.returns.list3 open requests
  • erp.orders.get · erp.stock.get2 exchanges possible, 1 refund
  • docs.searchwithin the return window
  • mail.draft3 drafts written

3 replies ready for review

Nothing is sent until a person approves it.

Brands we build for and run.Clients and partners we work with.

Case studies

HydraPak, Dometic, Klean Kanteen, Katadyn, Elkline, Spectra Watermakers, Regenliebe, nexpert, 24ocean, Strandsturm, nFusion

What we do

One company. Four disciplines.AI, automation, data, software.

We build the agents, automate the processes, engineer the data and ship our own products. Everything runs in our own operations before a client sees it.

Agents that take over real tasks with access to your tools, AI applications we develop ourselves, and local models that keep data inside your company.

AI at BrandUp

Business processes that run end to end without a person copying data between systems – with monitoring, retries and a log for every step.

How we automate

Pipelines that bring shop, ERP, marketing and finance data into one model – the base for reporting, forecasting and AI.

Data engineering

Our own products, built and supported in-house: Index AI, Section Store and Combined Listings – BrandUp on the Shopify App Store.

See the products

New · Local AI

AI that runs in your building.We build the hardware and keep it running.

For data that must not go to a cloud service: models on hardware you own, from a unit on the desk to a GPU rack. We size it, build it, install it and look after it.

Open the builder

2 × Lenovo ThinkSystem SR675 V3

Two GPU servers with eight RTX PRO 6000 SE cards each, switch and battery backup.

AI memory
1536 GB
Graphics
16 cards
  • Prompts and documents never leave the company
  • Real hardware, sized for your workload
  • Built, installed and maintained by one team
Build your system

How it fits together

One system underneath.Not five tools taped together.

Every source flows into one data model. Agents and workflows work on top of it, and what they do lands back in the systems your team already uses.

Example 01A customer writes in. An agent reads the order and the stock and drafts a reply for approval.

Sources: Shop, ERP, Inbox, Calendar, Ads.

Layers built and run by BrandUp: Agents (take over tasks, with tools and approvals); Workflows (run processes end to end, monitored); Data model (one source of truth for every system).

Outcomes: Replies drafted, Shipments created, Reports updated, Forecasts refreshed.

How we build

Engineering first.Everything else follows.

Every tool we sell runs in our own operations first. That shows in the stack, the release rhythm and the changelog.

pipelines/sync-orders.ts
1// nightly 03:00 - warehouse -> Shopify -> Klaviyo
2const orders = await plenty.orders.since(lastRun, { window: 'month' })
3for (const order of orders) {
4 const synced = await shopify.orders.upsert(mapOrder(order))
5 await loyalty.award(synced.customerId, points(order))
6 await klaviyo.track('Order Synced', { order: synced })
7}
8await monitor.ok('sync:plenty', { count: orders.length })
cron · nightlyType check gates every deploy
The stack
  • Shopify
  • TypeScript
  • Next.js
  • React
  • Node.js
  • PostgreSQL
  • GraphQL
  • Liquid
  • Tailwind CSS
  • AWS
  • Vercel
  • Docker
Shipped recently
  • Nov 13Now on the Shopify App StoreDigital Product PassportNew
  • Nov 12Deals are free on every planSection StoreImproved
  • Nov 11AI Sections for the Enterprise PlanSection StoreNew

TypeScript end to end

Apps, dashboard, pipelines and themes share one language and one type system. A type check gates every deploy.

Shopify APIs, not workarounds

GraphQL Admin API, webhooks, theme app extensions and the Partner API – the way Shopify intends apps to be built.

Our own data platform

PostgreSQL with hourly catalogue syncs and a unified order history across Shopify and the warehouse system.

Automation with monitoring

Scheduled pipelines with alerts and rollback paths. If a sync fails at 3 a.m., we know before the merchant does.

Infrastructure we run

AWS for data, Vercel for the web, Heroku for app backends. EU regions where the data requires it.

Small releases, weekly

Themes deploy through GitHub sync, apps through staged rollouts. No big-bang launches, no six-month waterfalls.

How we work

Three thingsyou can hold us to.

01

Fixed quote, fixed scope

After a short call you get a written price and timeline. No hourly surprises, payment in milestones.

02

Engineers on every project

No account-management layer between you and the people who build. Weekly status, one point of contact.

03

Your systems, your accounts, no lock-in

Code, repositories, accounts and data belong to you from day one. We work inside your accounts and can be removed any time.

BrandUp Factory GmbH, Hamburg – founded in 2019, Shopify Partner. Engineers on every project: no account-management layer between you and the people who build, run and support the software.

Company

Start a project

Tell us what you wantto build or automate.

One message is enough. We come back with a short call slot, a straight answer on what we would do – and a written fixed quote afterwards. No obligation.

  • Answer within one business day
  • Fixed quote after a 30-minute call
  • Talk to an engineer who works on your project
info@brandupfactory.com

Tell us about your project

An agent, an automated process, a data platform or an app – a few sentences are enough. You get a written answer within one business day.

We use your details only to answer this inquiry. No newsletter, no sharing. Our privacy policy applies.