Systems and power
How algorithms, incentives and institutional choices shape what people see and the opportunities they receive.
Ideas and research
Look past the headline. Examine the system.
My work focuses on the point where technology meets people and institutions. The aim is not to predict every future. It is to ask better questions about the choices being made now.
Work and employment
What technological change means for hiring, skills, productivity, job quality and the human experience of work.
Governance and trust
How leaders can turn broad principles into evidence, oversight and accountability that stand up in practice.
Selected work
Research and analysis
Published through H.A.I.R, Eunomia HR and independent research channels.
01Structural Properties and Systemic Risks in LinkedIn’s Modern Recommendation StackA research monograph examining how retrieval, ranking and notification systems can create unequal visibility, including through proxy and intersectional bias.02LLM Reality Check: AI Résumé Screeners are Fast, Fluent and Fundamentally UnreliableThree leading models, 109 anonymised CVs and 300 screening runs. The experiment found just 14% overlap between daily top-ten shortlists.03Disposition Data: The Asset That Became a LiabilityWhy historic hiring outcomes can encode the decisions organisations should be questioning, not reproducing, and what that means for AI governance.04The Burnout Trap: Why “Saving Time” with AI Might Be Breaking Your TeamA challenge to the assumption that every automated task creates capacity, with a closer look at cognitive load, pace and the shape of good work.05The Agentic Maturity LadderA five-stage model to help leaders distinguish basic automation from genuine agentic capability and ask better questions of technology providers.06Algorithmic Suppression: Is LinkedIn’s Algorithm Biased Against Women?A legal and technical analysis of how proxy signals in recommendation systems can reproduce unequal outcomes without using protected characteristics directly.