Your subscription could not be saved. Please try again.
Your subscription has been successful.

Newsletter

Subscribe to be notified when a new article is uploaded

Welcome to FinnLearns! This site is dedicated to sharing the various projects I’ve created and the lessons I’ve learned in doing so. But before you start browsing, I’d like to introduce myself.

Ever since I finished geometry when I was 15 years old I’ve been enthtralled by math and science, both for their own sake and because of their novel applications to technology. I’m especially interested in the application of math (especially probability) to the simulation of physical processes, and hence I am currently a junior in applied mathematics at CU Boulder. I’m also equally interested in math for its own sake, and thus I’m minoring in pure math. Finally, simply to satisfy my interest in the field, I’m also minoring in history.

In addition to my coursework, I also play guitar and have performed across Kansas and Colorado with my brother as an acoustic duo. To check out our work you can visit our YouTube channel which contains all of our music. Besides playing gigs for restaurants and private events we’ve performed multiple shows for retirement homes, fundraising events, a city-wide music festival in the Kansas City metro, and even a wedding!

Below are a few other personal achievements and fun-facts:

Statement on AI Usage

Anyone who is visiting this site is likely to keep somewhat abreast on the current situation with the proliferation of large language models (LLMs) in STEM. As this website is meant to showcase my projects, thoughts, and ideas, in the interest of transparency I believe it is vital to address how I use AI to augment this process and how you can expect AI to have contributed to the content on this sight.

First, my philosophy on AI adoption: it is here, it always will be, and it is therefore of utmost importance to set clear personal boundaries for the tasks in which you’re comfortable allowing AI usage. As the last year has shown, these boundaries must be fluid as the pace of technological development has taken all but the most hopeful optimists by surprise. Currently, I’ve personally set a clear boundary against using AI as a writing tool. Perhaps it’s because my education heavily emphasized writing, or because I naively view writing as something intrinsically human and reflective of the individual, but I’ve never felt the need nor the urge to adopt AI usage for penmanship. You can therefore expect every word on this site to have been personally typed by yours truly.

But besides writing, the only task in which I’ve set a hard limit against adopting AI is assisting with homework. When I’m working on projects, I feel that AI can be a great boon to augmenting one’s understanding of a problem. This is especially true in coding, where searching through poor documentation or snarky Stack Overflow threads could take up valuable time otherwise spent on more interesting activity (at least in my opinion). Certainly, asking AI for the best Numpy method suited to a certain task isn’t inherently inferior to grinding through a web search. However, there is a clear difference bewteen someone taking the mental leap of understanding and internalizing why that method is better as opposed to someone who craps out the AI’s answer into their code. In ye olden’ days, one would naturally take the time to understand the new method they’ve found: first, they would have to think deeply about the problem they’re trying to solve so that they can spot a solution; second, the payoff to your arduous search was the realization of how the new method works; and third, you would have to understand the method to properly implement it.

The first use case for AI that adopted was eliminating searching through documentation (at least for well-established libraries). I think that the above exmaple, however, can be extended to virtually any domain, whether it be coding, math, or any other conceivable problem. AI, like any process-optimizing tool in the past, should be used to augment understanding; however, no tool has been so adept at so many fields and tasks at once. With these thoughts in mind, this is how I approach:

  1. Identify a problem, and think about it using minimal outside resources. Identify gaps in understanding, first-pass implementation strategies, or anything else of this sort specific to the problem at hand.
  2. Perform a traditional, archaic web search for resources addressing, verifying, or proving the open questions fielded last step.
  3. Try a coding implementation or verify that the found sources actually answer the posed questions. If not, perform additional web searches, think more, or query AI with the explicit goal of understanding and verifying its output.
  4. Rinse and repeat step 3…

Hopefully this brief diatribe is an informative look into the processes I use to create the projects listed on this site. Even though the current pace of AI development could make the capabilities of the systems I’m currently writing about seem obsolete, for the goals of this site I believe these are practically immutable standards that I’ll be heeding.