Hey there!

I’m Elo, a PhD student in Mechanical Engineering at Louisiana State University, and this is where I’ll be sharing my thoughts on AI, computational science, and the occasional chaos of research life.

What I’m Working On

My research sits at the intersection of computational fluid dynamics and machine learning. Specifically, I’m building reinforcement learning agents that learn to intelligently optimize expensive numerical simulations. Think of it as teaching an AI to pick the right tool for the job, except the job involves simulating explosions at microsecond timescales.

The goal? Make combustion simulations faster and more efficient without sacrificing accuracy. meaning less time waiting for simulations and more time doing actual science.

What You’ll Find Here

I’ll be writing about:

AI & Machine Learning: Practical experiences with reinforcement learning, multi-agent systems, and applying ML to real engineering problems. Expect discussions on PPO, reward shaping, and why your policy sometimes decides to collapse spectacularly.

Computational Science: Numerical methods, scientific computing, solver development, and the art of making code run faster. I’ve built tools for chemical kinetics integration and worked extensively with CFD simulations.

Research Experience: The reality of PhD life, successful experiments, failed approaches, and lessons learned along the way. Also, thoughts on academic publishing, collaboration, and navigating the research landscape.

Startups & Innovation: Insights from working with INTERPOL on AI systems for child protection, internships in oil & gas optimization, and thoughts on translating research into real-world applications.

Tools & Workflows: Software I use, coding practices, and technical solutions to problems I encounter. Recent work includes Python package development, Docker deployments, and GitHub Actions automation.

Why This Blog?

First, to document my journey. PhD research is full of dead ends, breakthroughs, and lessons that are worth capturing. Second, to share knowledge. I’ve benefited immensely from others’ blog posts and technical writeups, this is my way of contributing back.

Also, let’s be honest: explaining your research helps you understand it better. If I can’t explain why my RL agent made a particular solver choice in a blog post, maybe I don’t understand it as well as I thought.

What to Expect

Posts will range from technical deep-dives (implementing custom RL environments, debugging numerical instabilities) to higher-level thoughts on research directions and career decisions. Some posts might be tutorials, others might be “here’s what I tried and why it failed spectacularly.”

I’m not aiming for perfection, just honest, useful content about the intersection of AI and computational science.

Let’s Connect

I’m always interested in discussing AI applications in scientific computing, reinforcement learning challenges, or anything related to energy optimization and combustion modeling. Feel free to reach out if you’re working on similar problems or have thoughts to share.

Thanks for stopping by. More posts coming soon.

— Eloghosa

Currently fighting with policy gradients and stiff ODEs in Baton Rouge, Louisiana.