AI Automation
AI Automation
Systems that take repetitive manual work off a team's plate, built with n8n, LLM APIs and the tools a business already uses.
AI CV Screening with Evidence-Based Scoring
Applications for one role arrive from three or four different places, nobody has a single list, and there is rarely anything written down explaining why a candidate was passed over.
The drop-in. For an agency that already has an intake process and wants the screening step automated.
Recruitment Operations Pipeline
Screening is the part everyone automates and the easiest part. Defining what good means for a role, keeping a record behind every rejection and chasing the calendar all stay manual.
The rebuild. For an agency rethinking the whole flow, from requisition to booked interview.
How I work
Every build starts with the process as it runs today, not with a tool. Who touches the work, what they read, where they sit waiting on somebody else. Once that is written down, the automation is whatever is left over.
AI stays narrow. A model is good at reading a document and forming a judgement about it, and bad at deciding what happens next. So the model reads, and ordinary code handles the arithmetic, the thresholds and the routing. That way a wrong answer is a bad score, not a booked meeting.
Nothing goes out to a customer or a candidate without a person approving it. Text gets drafted, a human sends it. Where the system is not confident it says so and escalates, rather than picking an outcome and hoping.
Have a process worth automating?
Book a free 30 minute call. Talk me through the steps your team repeats every week and I will tell you what is worth automating and what should stay manual.