Austin Attempts: Timely Journey

The only things provided by this puzzle are the Title: “Timely Journey” the below image, and a PDF copy of the paper seen in the image. I also checked the webpage using Chrome DevTools Inspect Mode to see if any hidden text was present but did not notice anything unusual. Therefore my assumption is that everything needed to solve this puzzle is contained withing the image and the title is most likely a hint. ...

January 6, 2026 · Austin Guevara

Austin Attempts: Robot Javelin

Jane Street’s Robot Javelin Competition Unfortunatly, my entry for this puzzle was incorrect. However, Here is my inital logic and thought process along with notes for where I went wrong. Robot Javelin There is a group of robots that compete to throw a javelin as far as possible in head-to-head (1v1) competitions. The Robot that submits the longest throw wins the competition. Rules Each competitor will throw their javelin once. Based off the distance of their first throw the Robot will then decide if they want to keep that distance or rethrow. If they rethrow they are required to submit the distance of the new throw. The robot has no information on opponent’s throws Givens A robots throw is a random number drawn uniformly between $0$ and $1$. The game has reached Nash Equilibrium and therefore all Robots are using the same strategy. Nash Equilibrium Strategy As stated in the rules each Robot’s actions are independent of their competitor. This means that this strategy is purely based on increasing the probability of maximizing ones own score. Because the first throw is purely random and required, all strategies would throw the first javelin then the strategy would be whether or not they throw the javelin again or keep your first distance. In a unifrom distribution of distances a distance of $0.5$ is the mid midpoint. This means that a new throw has a $50%$ chance to be less than $0.5$ and a $50%$ chance to be greater than $0.5$. Based on this the Robot should only ever rethrow if their is a greater than $50%$ chance that the rethrow will improve their score. ...

January 5, 2026 · Austin Guevara

Austin Attempts: AOC2025

This was my second year diving into Advent of Code (AOC), the annual programming puzzle extravaganza. For the first time, I managed to complete all puzzles within the event. The shorter 12-day format this year (down from the usual 25) probably helped! Here’s a quick tour of my journey through AOC2025, day by day. Day 01: Secret Entrance A gentle start. The challenge was rolling a “lock” from 99 to 00 and back—trickier than it sounds if your language’s modulus operator behaves differently. Once I got that right, I just needed two counters: one for stops on 0, and one for passes over it. ...

December 1, 2025 · Austin Guevara

Austin Attempts: Self Hosting AI

AI Hype 🚀 Over the past year AI hype has gotten out of hand. With the cultral populatiry of OpenAI and the general leap in sucess the space has seen in the past couple of years this hype isn’t unfounded. People outside the tech industry are increasingly recognizing how transformative these technologies can be, potentially displacing jobs on a scale similar to the Industrial Revolution. This is causing a rush by practically every industry to be at the forefront of these developments. I, on the other hand, have been extremly wary of these tools. I tried my best to avoid them unil Nordic SemiConductor added an AI assistant to their Dev Zone. These microcontroller companies have a ton of resources however they are spread out through forums, app notes, white papers, and data sheets. This makes it a pain to locate and process without an FAEs help. This was the first time I used AI and felt it was genuinly useful. The tool would find a relevant response and do its best. It was often wrong, however, it was close enough that somone familiar with the technology could peace together a solution. ...

February 7, 2025 · Austin Guevara

Austin Attempts: Writing a Blog

Why did I start this? Honestly, I blame this entire journey on two people: Beau and Chris. As 2024 came to an end, I decided to focus on improving my coding skills. For the first time ever, I participated in Advent of Code and to make it even more challenging, I completed it entirely in a language that was completely new to me—Zig.1 This sparked a series of tangents, including learning RayLib, attempting to create a browser-based game, and eventually diving into Go to build an HTTP server. ...

January 8, 2025 · Austin Guevara