Welcome to my digital garden!

My name is Ed, I am a researcher in alignment and quantitative finance.

In this garden you will see my daily routine in solving one of the most important problems of our time and sometimes side-stories.

In case you are looking for a researcher to work with or discuss ideas, you can feel free to write me here. I have wide range of interests in alignment.

List of finished ideas:

• Foundation white-paper: https://github.com/kapedalex/Foundation. Addresses open/closed AI governance and proposing an economic incentive framework for alignment.
• Cyclic ablation paper: https://arxiv.org/pdf/2509.25220. Emphirical proof that iterative ablation the concept of deception is ineffective, +1 to manifold hypithesis.

List of current works:

• Developing a mechanistic unlearning techni que leveraging geometric similarity via Gromov-Hausdorff distance methods to detect and suppress hazardous subspaces.
• Defining agentic preferences. Have done some work, still not sure will SPAR team be helpful or better to work alone.
• Examining self-led agents behaviour. Quite related to previous one. Bigger part of work these days will be related to Agency.