Hi, I'm Lu. I research AI fairness and safety, and I lead complex programmes to delivery.
I want AI to be built and deployed safely, fairly, and for genuine public benefit.
I bring 13 years of experience delivering complex digital programmes across government and SaaS scale-ups, and I'm currently completing an MRes in AI-Enabled healthcare at UCL, researching causal methods for mitigating unfair bias in clinical AI.
I'm looking for operational roles in AI safety and public-interest AI, where governance, research, and delivery have to actually meet.
Operations & Program Management
I get things done. Across whatever operational problems need solving.
I've stood up governance from scratch when none existed, coordinated teams across disciplines and organisations to launch national infrastructure, sourced and hired talent under pressure, and managed the delivery of complex programmes within tight timelines.
Along the way I've learned that the hardest part for teams is usually figuring out what actually needs doing, and why. 13 years across government and startups taught me the same lesson everywhere: understand the real problem, then go solve it.
AI Safety & Governance
I write about AI ethics, safety and governance issues for a general audience.
I am particularly interested in building soceital resilience and preserving human agency.
My writingAI Fairness Research
My MRes thesis (UCL) develops a Structural Causal Model and a novel architecture — CEVAE-HE — to separate legitimate biological variation from unfair sociological bias in clinical data. I validated the framework across controlled synthetic populations and a real clinical dataset, with a close focus on where causal bias-mitigation methods hold up and where they don't.
Alongside my MRes at UCL, I independently built an agentic pipeline to generate synthetic data for fairness research.
My research