Hey everyone, I'm Aneesh.
Today, I’m launching this developer log as a personal accountability challenge. From this week onward, I am committing to publishing at least one technical post every single week (and more than that if I run into breakthroughs or major roadblocks worth sharing).
My goal is to document the raw, unfiltered engineering journey, the bugs, the design choices, the math derivations, and the late-night simulation wins.
What I'm Building: Kaggle Orbit Wars
Right now, my primary focus is the Kaggle Orbit Wars simulation challenge. It's a brutal 2D real-time strategy environment where you have to conquer rotating planets, navigate gravitational paths around a central sun, and optimize fleet trajectories. With about a month left until the final submission deadline in June, I am aggressively iterating on my agent's state estimation and decision-making logic.
What I'm Learning: UC Berkeley's CS 285
To back up my practical work with deep theoretical foundations, I am currently working through UC Berkeley's CS 285 (Deep Reinforcement Learning) course. Shifting from hard-coded heuristics to understanding advanced policy gradients, Q-learning value functions, and model-based RL is completely reshaping how I think about designing autonomous agents.
Why I'm Doing This
I'm skipping the corporate noise of traditional social media. This space is going to be my open-source lab notebook. If you are also grinding through Orbit Wars, studying RL, or building autonomous systems, follow along or drop a comment—let's build together.
See you next week for the first deep-dive update!
Top comments (0)