AI Workflow Automation

Connect your tools and let AI handle the work in between

For teams whose day goes on moving information between systems. I build workflows that connect your apps, including through MCP, and use AI where a person used to decide.

Book a free 30-min call
System architectureApplied AI
01Audit
02Build
03Run

Your tools don't talk to each other

Data arrives by email, a person checks it, retypes it into the CRM and reports on it in a spreadsheet. Classic automation breaks as soon as a step needs judgement. AI can make that judgement, as long as it's wired into the right place.

  • CRM
  • Email
  • Internal tools
  • AI judge / classifier
  • Automated actions
  • Human review
Your tools feed an AI judge that decides what happens next, while a person reviews the edge cases.

What I do

  1. Connect your tools

    I integrate cloud apps with internal and on-premise systems through their APIs and MCP servers.

  2. Add AI where judgement is needed

    AI classifies, extracts, scores or approves items, and flags anything uncertain for a person.

  3. Keep it running

    Workflows run in the background, with retries, logs and alerts when something fails.

Example use cases

  1. Inbox triage

    Incoming emails and forms are classified, summarised and routed to the right person or system.

  2. Lead enrichment

    New leads are enriched with third-party data and scored before they reach your sales team.

  3. Document processing

    Invoices, orders and contracts are read, checked and entered into your systems automatically.

Proof

How it works

  1. 01

    Pick the workflow

    We choose one workflow with a clear volume and cost.

  2. 02

    Map the steps

    I document every step, system and decision in the current process.

  3. 03

    Build and connect

    I build the workflow, connect your tools and test it on real data.

  4. 04

    Measure and extend

    We compare the time saved with the old process, then automate the next workflow.

Frequently asked questions

What is MCP?
The Model Context Protocol is an open standard that lets AI models use your tools and data in a controlled way. I use it to connect AI to your systems without building a separate integration for every model.
Can you connect our internal or on-premise systems?
Yes. I connect internal databases and on-premise tools as well as cloud apps, through their APIs or a small connector where none exists.
What happens when the AI isn't sure?
Uncertain items are flagged and sent to a person. You decide the threshold, and you can tighten it as your confidence in the workflow grows.

Automate your busiest workflow

In a free 30-minute call we'll pick a workflow and sketch how it would run with AI in the loop.

Book a free 30-min call