Introduction

I’m a product and design engineer as well as an AI safety researcher. I currently work on contract for Anthem, one of South Africa’s largest renewable energy providers in analytics.

Alongside ML engineering work, I engage in 15+ hours a week on technical AI safety. My research focuses on mechanistic interpretability and AI personas in the context of alignment. I approach my theory of change through field building, working with strong talent. I’m able to translate between specialisms and aim to bring technical, governance and execution closer together. I have additional interests in working with enterprise on governance, stable transition, and future of work.

I use a combined skillset of design, ML, business and project management to solve problems. My programming skills cover data science and explainable AI, closely following the research published by Anthropic.

I founded and ran a product engineering company, Story Springs, in the UK for four years. Clients included Greenpeace, Harper's Bazaar & Tarzan. Before this I worked as a business consultant for SMEs in the creative industries, specifically business structure and revenue.

In 2024, I received funding to participate in a ten month Google Quantum AI training programme, and was a member of an informal Google development group for error handling tools. I am part of a weekly AI safety reading group, regularly attend AI safety conferences and have participated in Apart AI safety hackathons, including technical governance and AI control.

My background in design engineering helps me to work in interdisciplinary teams and projects which require drawing from a wide range of skills in order to build effective and intuitive models, products and frameworks.

I have been a regular industry speaker in London, talking at the Media & Production Technology Show, Starting Pixel and the Smalls. I participated in several UK & international fellowships.

I hold a Bachelor of Commerce from UCT and an MSc in AI and Data Science from Goldsmiths, University of London.

Skills / Core Competencies

• Predictive Modelling & Statistical Analysis: Python and ML Libraries, R, MATLAB, Object-Oriented Programming, Cirq, Qiskit, Stim.

• Cloud Computing: Vercel, Supabse, Railway.

• Mechanistic Interpretability: concept vectors and vector steering.

• Technical AI Safety: alignment, control, research, personas.

• AI Policy & Governance: ethical frameworks, regulatory compliance, risk decomposition, and risk mitigation strategies for responsible deployment. Interests: compute governance.

• Strategic, Business and Project Management

• Feature Selection & Problem Framing: Pre-processing real-world data and problems to be machine readable.

• Information Architecture: Organising complex data sets so they are legible and useful to decision-makers.

• Human-Computer Interaction (HCI)

• Design Engineering: combining design and engineering to bridge the gap between design intention and code.

• 3D Design & technologies: AR/VR, motion and audio control, movement data tracking, image tracking, virtual production.

• Visual & UX Design: Designing how data flows through a user interface