About me

I'm a researcher working on AI safety, with a focus on understanding agency, both in AI systems, and to understand how powerful AI will impact human agency. Currently pursuing an MPhil in Applied Mathematics at the University of Cape Town, where I'm studying the application of mechanistic interpretability to deeply understand RL models. I also have strong interest in evaluative frameworks and creative ways that we can learn to measure and understand LLMs.

I am also passionate about teaching and growing the field of AI safety, and co-founded AI Safety Cape Town to contribute to that effort.

Outside of research, I draw inspiration from Buddhism philosophy, Stoicism, and Kantian ethics in thinking about how to make meaningful contributions to the field of AI safety. To recharge I spend time playing beach volleyball, watching anime, reading epic fantasy/sci-fi and hiking.

Philosophies

I am a huge fan of stoicism. From Marcus Aurelius, to Ryan Holiday, to Seneca. There's something incredibly satisfying about absording these ancient ideas and applying them in life. It's a really powerful operating system in general.

Taoism is also one philosophy that I have to mention. Taoism (or Daoism) seem to capture a deep truth about the world. My favourite texts in this space are the Zhuangze (chapters 1 to 7) and the Tao Te Ching (ideally the most unabridged version you can find). I look forward to the day when I can truly stop seeking external validation and just exist for the sake of existing itself.

I love a lot of what is contained in the essence of the Buddha's teachings. There is a rich tradition of writing and ideas in the Buddhist tradition and I connect very strongly with the ideas of how to conduct oneself and the fundamental world view of the Buddha which is that we manufacture our own reality based on a self-serving world view that often involves a lot of unneccessary suffering and through this we become unable to make clear sense of our reality.

I love rationality, scientific thinking, mathematics, art, music, good design, and being around plants.

Biographical info

My teenage years were spent at an interesting school called Cedar House, which took a broader view than most of what a school should look like, and which gave a lot of freedom and autonomy to its students. By my estimation, it was a very good education. I moved on from there to Rhodes University. The people there were very unique and quite divergent thinkers. It's difficult to describe exactly what sets them apart, but in general I found them very friendly, wholesome, and a little bit whimsical. The education was decent, and the town and the feeling of being in an isolated community led to some really fun years playing a lot of Dungeons and Dragons.

For a long time I played video games, or distracted myself on the internet instead of pursuing things which I felt I really needed to be doing. I've made many changes in recent years to how I approach pleasure and entertainment, as I found pursuing pleasure for its own sake was ultimately totally unsatisfying. Somehow, I still enjoy everything that I do, but those activities are not centrered purely around pleasure, but scratch some other, deeper itch.

Meditation

I experienced a massive shift in 2019 when I gained new mentors such as Tim Ferris, Derek Sivers, and read books like Awareness by Anthony De Mello, which gave me a better framework for understanding spirituality. It shifted my thinking from trying to numb the unpleasant sensations and constantly try to experience positive ones, but rather to look beyond the structures of living that had held me for so long. I began to realise that happiness was a state of mind that I was responsible for creating. That my interpretation of what happened around me determined my response. I realised the possibility of mastering my own mind.

At the beginning of 2020 I began a serious vipassana meditation practice following a 10 day course with S.N Goenka. I now work hard to maintain a strong daily practice. I don't think I could trust myself to be a good human without that practice to ground me in my life. Meditation helps to not cause harm in the world and it makes it easier and more natural to want to serve others.

A huge part of my self development journey as well as my meditation journey have been about trying to soothe the symptoms of the ADD. I have found that unless I take it seriously and create an environment which supports me then my life becomes very difficult.

What I'm working on

I love code, computers, technology, and AI. Currently I am working on technical alignment research and I am dedicating my career towards AI safety generally. I am not necessarily terrified of inevitable extinction from AGI, but I do think that the chances are uncomfortably high that if we build an AGI it will have catastrophic consequences, and I think we might stumble into that technology much faster than we expect.

My current work aims to focus on instilling moral values into RL agents, and trying to make RL agents more interpretable through techniques developed in the natural language processing domain. I am also very interested in developing evaluations for testing alignment in large language models.

Reading and Writing

I am a passionate reader and my plan for early retirement involves switching to being a fantasy author. The first series which I remember being my favourite was the "His Dark Materials" trilogy. My love of fantasy and sci-fi stories was and still is an unquenchable fire.

I have since spread out and now I read from a broad spectrum, but I find it easiest to get obsessively stuck reading fantasy and sci-fi.

Acknowledgments

This website's format is largely based on Derek Siver's website. He's such an inspiration to me, and shifted the course of my life through his book recommendations and thinking.