Process automation with AI

We map one process end to end, then automate the reading, checking and re-typing in it, with people kept on the decisions.

Invoice flow
3 of 4 on
  • Read supplier invoicesPDF and email
  • Match to purchase ordersPrice and quantity
  • Post matched invoicesInto Sage
  • Pay without approvalA person signs off

The problem we solve

Most slow work is not hard work. It is reading an invoice, finding the matching order, copying numbers into another system and chasing someone for a signature, hundreds of times a week.

We sit with the people who do the work, map the process step by step and automate the parts a machine does well. Every automated step is logged, and the decisions stay with your team.

Typical engagement

  • Process map (1–2 weeks): every step, every system it touches and where the time goes.
  • Automation (4–10 weeks): the first automated flow in production, measured against the old one.
  • Next process: the same method on the next bottleneck, with your team running the first one.

What we deliver

  • A step-by-step map of the process as it runs today
  • The steps worth automating, ranked by hours saved and risk
  • Automations for document reading, matching, routing and data entry
  • Approval points where a person signs off
  • A before-and-after measure of time per case

Tools we use

We pick tools to fit your systems and your team, and these are the ones we reach for most.

  • Python
  • OpenAI
  • Anthropic
  • n8n
  • Airflow
  • PostgreSQL

Related work

  • Automated testing platform

    An internal app that automates testing workflows, stores results and uses machine learning to predict where tests will fail.

  • Property auction & asset management

    An auction and asset management platform with live auction workflows, asset listings and reporting, fed by ETL data pipelines.

  • Counselling platform MVP

    A therapy-adjacent platform with affirmation tracking, interaction logging and a recommendation engine that personalises each user's journey.