I’m an experimentalist, autodidact, and forever an amateur. The place you find yourself in has become my personal space for learning, research, software, nanotech, and unfinished theories. Perhaps one day it will turn into something more. Maybe in the distant future. Or maybe not, c’est la vie.
A life without learning is boring to me, it means I’m likely not challenging myself nor growing as a person. I find many people repeat the same day for the rest of their lives but that’s never been my style. I find meaning through the contrast of the extremes I’ve lived through in my life. What makes me feel fulfilled at the end of a day is learning new things and having accomplished lots wether that’s in software development, my amateur research, or spending my time learning about niche scientific/technical subjects. The what isn’t as important as the how. I take what I spend my time on seriously with a few exceptions, pursuing high-quality work is important to me. I don’t like half-assing things and rarely do so unless it’s for fun or for something to be thrown away.
Why the name Graph Society?
I’m addicted to network graphs. Maybe the domain gave it away.
Change the graph, change the world.
~ a personal motto
Human culture, cognition, and our minds are often best studied as networks. From molecular networks, to neural networks, to networks of association accumulating knowledge itself, to animals organizing into social networks, we’re practically composed of networks and all of them can be manipulated and modified if you learn how to do so… ʚїɞ
About me
Lately I’ve been learning about how to grow nanonetworks, improving my math/physics/chemistry skills, learning, researching, and writing software.. My work draws from computer science/software engineering, mathematics, network science, physics, past experiences, my taste, and the intrinsic intuition I have for how to direct my time and energy. My main focus at the moment is conducting reservoir computing experiments and upgrading some of the skills I’ve let stagnate.
I dropped out of my first year of highschool and as a young adult I was kicked out of my parents place. I spent my time working in a warehouse part-time while continuing to study software engineering/computer science on the side. Software has always had a place in my heart, at 15 was my first venture into the world of code where I started learning C++ from a book by Bjarne Stroustrup.. Later on I broke into the industry professionally at an ad-tech company where I got my first taste of seeing AI in action and working in an “Agile” environment. I had multiple job offers at the time. That was way back in 2020/2021. At my first job one of my coworkers mentioned to me it was like I already had many years of professional experience. I was always unsure of my skills before that comment so thanks Claudio!
My time working in software influences how I now perform research today. In a way, what we spend time on is who we become. And in software, reproducible outcomes are important.
Doveryai, no proveryai
Reader beware!
Trust, but verify.
There are gaps in what I know. I’m also just some person on the internet. I know some things pretty well and other things not-so-much, with a pretty vast in-between. By the very nature of learning in public, some things here will be straight-up wrong, if not at least not as correct as they could be. Even with that said I aim to try my best to be accurate in my writing. I’ll publish my data, methods, and observation so if wanted you can try to verify my work too. In fact I’d really appreciate if you did just that. I’ll likely store such things somewhere online, perhaps on my Github. We’ll see…
I use agentic AI/LLMs to assist in my process, for example:
- Writing code.
- Occasional brainstorming.
- Finding unsolved problems.
- Searching for better algorithms.
- As a search engine for various vague queries.
- Pulling research literature by keywords.
- Looking for insights across various domains I’m both familiar and unfamiliar with.
- etc…
Even though I use LLMs as an assistant all the writing found at graphsociety.org is of my own. I still verify the important parts of what AI completes, and I question it’s conclusions when needed. I mostly focus on performing the deeper intellectual work of direction and guidance. Most of my progress comes from acquiring information, understanding it, and double-checking my assumptions and data acquisition procedures where each round informs the next. A serendipitous feature of this is that the same abstraction often appear over and over again in different contexts. This allows me to reuse old battle-tested tools alongside building new ones tailored to my context. And who nose, maybe if I’m lucky I’ll invent or discover something that outlasts me.