Links

My open-source code.

Books I liked and books I want to read.

I have an account but barely do anything there - I want to be more active on LW in the future.

My professional profile.

Technical research forum for AI alignment. Similar to LessWrong, but exclusively for Alignment

The main online home of the rationalist community and a major hub for AI safety discussion. Mix of rationality writing, alignment research, and general intellectual content.

Free intensive bootcamps that educate about AI safety. A good entry point for (not just) technical people who want to move into the field. I wrote about my experience with them <a href="https://www.lesswrong.com/posts/qovWG7EmYBzcea9Mh/ml4good-ai-safety-bootcamp-experience-report">here</a>

Janus's essay arguing that base LLMs are best understood as simulators of characters rather than agents or oracles. An influential frame for reasoning about what current models actually are. Note that in today's models, post-training plays a large role as well, this essay is about base models. Nevertheless it shaped my thinking about LLMs.

Harry Potter and the Methods of Rationality, a fanfic by Eliezer Yudkowsky. An entertaining read and introduction to concepts of rationality.

Eliezer Yudkowsky's essays on rationality, epistemics, and reductionism. Very influential in the bay area and on LessWrong.

The rebranded Open Philanthropy. One of the largest funders of AI safety research and other high-impact causes, they also funded me :) If you need funding, check with them.

Excellent podcast, sometimes with AIS people, often general tech or history. I really dislike most podcasts, but Dwarkesh is a rare exception.

One of the earliest voices arguing that misaligned AI poses an existential risk. Much of the modern AI safety discourse traces back to his work. Polarizing figure, but extremely prescient & unedniably smart

Ex-mathematician in finance, now statistics professor. Focused on "fat tailed" distributions and how much of modern statistics does not work in real life. Polarizing figure, but always brings an interesting perspective. Very childish on Twitter :)

Scott Alexander's blog. Very broad in scope and overall a very clear thinker. He used to work as a psychiatrist and has a lot of background knowledge there, plus a strong interest in AI and politics, but stays data-driven and relatively objective.

Tim Urban's blog. Presents many topics in an entertaining, illustrated way. Epistemics are not always the best (e.g. the post series on Elon Musk's companies), but generally worth reading.

Don't always agree with him, but he generally has interesting takes.

Dr. Aaron Horschig's YouTube channel on lifting mechanics, mobility, and rehabbing common gym injuries. I find him very useful for lifting and training advice.