AI & Automation · Forecasting & data
Forecasts a planner can question, by product, store and day.
Three model families compared on your own history, served through an API your planning tool reads, with a plain-English explanation of every number.
- SKU × store × day
- 3 model families
- fine-tuned on your data
How it works.
- 01ETLfrom your SQL Server or ERP
- 02Analytics storewith data health checks
- 03Modelszero-shot and fine-tuned Chronos, TimesFM, NeuralProphet
- 04Forecast API
- 05Dashboardsplus "ask the forecast" chat
From case 18, built for a fashion-retail AI platform.
What else we do with your data.
Data pipelines
From your systems into one analytics store, with health checks.
Product-data cleaning
Classify, de-duplicate and clean images at millions of records.
Case 13.
Computer vision
Product recognition and defect checks.
Case 19.
Background removal
Car photos without the dealership background.
Case 10.
When the history doesn't help.
- A new product with no history
- It's flagged as low-confidence, and the planner sets the starting number until real sales come in.
- A promotion
- The planner adjusts and locks the forecast for the promotion period, so the model doesn't overwrite it.
- Missing data
- The data health checks flag the gap before the forecast runs, so a person sees it before anyone orders against it.
Forecasting and data work.
- Retail AI
Demand forecasting a buyer can question
3 forecasting model families compared · fashion and FMCG · 2026
- Market data
Cleaning product data for millions of e-commerce items
Millions e-commerce products · NIQ (Etilize) · 2009 to 2024
- Computer vision
Recognising luxury watches from a photo
4 luxury brands recognised · vintage watch shop · 2025
- Automotive
Car photos without the dealership background
Automatic background removal and placement · delivered
How it's priced.
- Sprint $490: a forecast on one product family from your sales export.
- Build scoped per data source.
- Software + AI team for ongoing work.
Prices exclude taxes and third-party costs (hosting, carrier minutes, AI usage), billed at cost to your own accounts.
See all pricesQuestions about forecasting.
How do you test it before it goes live?
On labelled test sets built from your own data, with automated evaluations, then staged UAT with the people who will use it. High-stakes actions stay behind approval gates after launch.
Can it run on our servers?
Yes. Open models can run on your own GPUs, so nothing leaves the building. We've done this for invoice reading, daily compliance checks and a receipt reader.
Does it replace staff?
It takes the repetitive part: first replies, data entry, routine checks. Your staff handle what it hands over, and they sign off each part before it goes live.
Send us one job.
Tell us what you need built. You'll have scope, price and dates in writing today, the work in 3–5 business days, and you pay only if you approve it.
Prices exclude taxes and third-party costs (hosting, carrier minutes, AI usage), billed at cost to your own accounts.