In the Spring of 2026, my team lead at the LIVE Lab asked me to reorganize the entire 3D art assets repository. The repository consisted of thousands of static meshes, thousands of textures, hundreds of HDAs, and an infinite number of Maya/Substance/Zbrush files.

So I got to work and began renaming assets according to our lab's standard naming conventions. But then I realized that this could end up taking up to 50 hours just renaming files and making errors along the way because I'm doing everything by hand. With this in mind, I decided to make a file-renaming tool to organize our repository! That way, I could not only do my task faster, but I could also give the tool to the team to make their lives easier with naming conventions for years to come.

The key features that I wanted were:

1. The ability to look through a folder and flatten the directory

2. Checking for proper naming conventions per file type (Ex. "S_" for static meshes (.fbx, .obj))

3. Counting assets with the same name and adding a "_#" suffix when necessary

4. The ability to identify texture type by scanning the name of texture files

5. The ability to redirect all texture files to a "Textures" folder

6. The ability to copy all HDAs into our separate HDA Library, and rename them based on what project they were for

   • I felt that this was an especially important task because it prevents us from doing work that other lab members had done in the past. Oftentimes, I found myself creating HDAs, and afterwards, my lead told me someone had made that 2 years ago. With this functionality, lab members can see every HDA that we've ever made!

So I got to work! I finished this project in about two weeks, which was significantly less time than it would have taken if I renamed every asset by hand. It was also helpful to my own personal skill development, as I hadn't touched Python since my Freshman year programming classes. This assignment was a great refresher and learning task!

Here is a display of the tool in action:

As you can see, I created my own custom UI as well. This was my first time ever creating a UI for my Python projects, and I created it in Qt Designer. All buttons and the title were created by me in Adobe Illustrator. I included a button where the user can select their desired folder, a dropdown where they can pick the project their asset is for, and a field to insert an asset name.

In addition, I found it useful to break up the buttons by file type instead of just having one button for all of it. That way, for example, the user could only rename Maya files if they wanted to, because the .fbx files had special naming conventions. Or they could only clone their HDAs into the HDA Library.

Here is a visual breakdown of the logic:

This was a super rewarding project at the end of the day! It helps my teammates and any future hires out with file organization. I hope to do a lot more Python projects like this in the future!

The Tree Generator is a tool that I made for a project at the LIVE Lab. We needed the ability to create a bunch of trees with ease for multiple different landscapes! Therefore, instead of opening up SpeedTree and creating dozens of different trees, I figured that I would just make a tool for that.

This HDA works by resampling points along a line and growing branches out from there. For this tool, I wanted to experiment with the "golden angle". The golden angle is essentially when leaves, branches, or other structures grow around a stem (~137.5°) to help them avoid directly overlapping, giving each leaf better access to sunlight, rain, and airflow!

There are a bunch of different controllable parameters for the growth of the tree (looking back, almost too many parameters).

Here is a breakdown of the HDA in use:

A functionality that I found especially important for this HDA was the ability for the artist to choose between hand-modeled leaves and leaf cards. We were debating between making the leaves a material and actually hand-modeling leaves, so I left that decision open and supported both options!

This procedural building was my first deep dive into procedural generation in Houdini. I had made HDAs, VFX, and simulations before, but had never dug deep into making a complex HDA. So I decided to do just that in a couple of months!

I made this project for both my Game Development and Creating Digital Environments classes. I decided to follow Anastasia Opara's Lakeside Houses tutorial series. It was a wonderful series and taught me so much about creating systems that work together. It was great for a medium-length project!

Here is a logic breakdown:

And here are a bunch of the seeds generated by the HDA!:

One of the biggest obstacles I ran into was with the roofing. Getting all of the different roof directions to mesh together was super complicated! What I ended up doing was booleaning any roof sections that intersected each other so that there wasn't any overlapping geometry!

All of the modules (except for the plants and the stairs from the Substance 3D assets library) were made by me for this project. To be honest, I had a little bit of scope creep going on for this project because I was so convinced that I wanted to be an environment artist as well at the time. So about two weeks of time was wasted creating and surfacing modules that never ended up being in the final product.

But all of it was so rewarding and taught me a lot! I'm so excited for my next procedural generation project where I get even more complex with my systems and hopefully not die from scope creep again.

Finally, here are some bonus renders:

Over the summer of 2026, I wanted to go into a kind of procedural generation I had only done from tutorials before: generation along a curve! I was inspired by Japanese tea gardens, and how they naturally transition from pathways to stairs to bridges. So I created a tool that takes the angle of the terrain below it between each point and the distance between each point and the ground to automatically place the correct path types along the curve!

Here is a quick video of the tool in action:

Within the tool are essentially three separate tools: One for the path, one for the short bridges, and one for the large bridges.

For both types of bridges, the planks are procedurally generated, as well as any accompanying geometry!

The stone path tool automatically places fences with breakup and variation alongside both sides of the path! In addition, the smaller cubes scattered alongside the path stones are essentially points where plants, pebbles, etc. could be placed procedurally.

I had a lot of fun with this project, and it served as a valuable learning experience on how to make different HDAs work together!

This project is currently in progress; check out my dev logs to see where I'm at!

For this project, I was responsible for creating the dithering shader, procedurally generating the town, and creating all of the boss fight VFX! I was also responsible for all environment layout and lighting.

Here is the trailer!

Play our game here!

(Card game VFX)

(Drunkard fight VFX)

(Priest fight VFX)

(Banker fight VFX)

(Dealer fight VFX)

I really loved this project because it was my first time making a large game with a team! We all had very different skill sets but worked so well together. In addition, we all got to be involved in the story-making and planning process. I can't wait for my next game like this!

Over the winter of 2025, I decided I really honestly just wanted to make the Traveling Cart from Stardew Valley in 3D (because it's so cool, obviously!). The goal of the project was to satisfy my craving for a 3D Traveling Cart and show my ability to take an asset from concept to 3D.

My first step was making a concept for the cart. It was relatively hard because the cart is only shown from the front in the game, so I had no reference for its dimensions. But we made it through and ended up with a snazzy concept sheet that I am quite proud of! I drew the concept in Clip Studio Paint, the software I used to make webtoons in High School back when I was obsessed with it.

Then came time to model the asset. The part of the modeling I had the hardest time with was the pink accent pieces on the front of the cart. I had never messed around with sculpting before, so it was definitely a fun exercise to learn something new!

One of the adjustments I made was with the width of the model. I felt that it was good in concept, but when I started making it in 3D, I found that it destroyed the look of it. Overall, I tried to keep the shape language stylized but clean. I'm very satisfied with the outcome, and honestly want to try modeling it again when I get a free weekend! (It was modeled and UV unwrapped in Maya)

After making the model, it was time to surface. I wanted to explore hand-painted surfacing for the first time! It was such a lengthy process and took so much trial and error, especially with the color palette. But I eventually figured out the style I wanted! This surfacing pass used exclusively base color and no other maps. It was done in Substance 3D Painter.

Next time I do a similar project, I would like to be more experimental in my modeling style and add more breakup/variation. In addition, I'd like to learn how to paint normals and incorporate other maps to match the hand-painted style of the base color map, since the effort of painting every single highlight seemed not entirely worth it!

But overall, I think that I gained a lot of valuable experience from this project. And definitely scratched that itch to make something from a game that I love!

P.S. The color palette theme for this website was based on this cart! We love jewel tones!

Spirit Street is a project that I completed with a team for my Creating Digital Environments class (VIST 372).

Monica Arce - Modeling & Surfacing

Hansen Han - Rendering

Me - Modeling, Surfacing, & Project Management

Valeria Saleh - Modeling & Surfacing

Miranda Avila-Alvarado - Modeling & Surfacing

Christabel Garza - Modeling & Surfacing

The objective of this project was to learn how to use USD workflows in Houdini to render a scene! We were heavily inspired by the stylized look of Spirited Away and wanted to translate the 2D elements of the film into a 3D stylized space!

I decided to break up tasks for each person by giving each person a building to work on. Since there were 5 buildings and 5 modelers/surfacers, it worked out fairly nicely! The only issue was the ability to match styles. Therefore, we had a meeting beforehand and decided which tools we were going to use and how we would go about our painting techniques. But naturally, each person will always have a different brushstroke style. But we decided to embrace the differences that remained after finishing the buildings!

I was entrusted with the green building on the left. Modeling the details on the front was definitely the hardest part.

I decided to only paint the front because any other work would have not been visible in the final render. Every single detail is hand-painted, including any grunge, highlights, shadows, etc. No other maps were used but base color. I really enjoyed painting the stained glass windows, specifically!

Here is the final product…

Tada!!! I can speak for all of us when I say that we are proud of how it turned out. The scene felt homey, whole, and filled with each of our personal styles. The most rewarding part of the project wasn't even making the asset; it was working with my teammates!

This project was made in my motion capture class with Michael Walsh at Texas A&M University. I took this class to expand my tech art skills into animation and retargeting. In this class, we learned all the ins and outs of MotionBuilder, and actually got to capture the motion in the volume ourselves!

After receiving the motion, I went through the process of cleaning up the bumps and imperfections in my character's movement. The Unreal character I decided to use was Paragon's Drongo, so I had to make sure that the motion didn't conflict with any of the gear he has on his body. In addition, Drongo is naturally hunched, so that had to be kept in mind when directing the actor while they were suited up.

My workflow involved baking each animation down to the skeleton and exporting them out as one FBX. This made it fairly easy to import into Unreal! All that I had to do was go through the retargeting menu and ensure the skeletons had the same bone hierarchy. From there, the rest of the project became camera work, environment layout, and rendering!

Here is the final product:

This project came in handy for a lot of the work that I was having to do at the LIVE Lab that semester. For a couple of months, my job essentially involved retargeting Mixamo animations to a custom character with a root bone so that we could use root motion lock in the game.

Overall, I learned so much about retargeting through this project and my work at the Lab! Would love to do it again, especially in a game context.

Over the winter of 2025, along with my Traveling Cart, I wanted to explore grooming! With my developing interest in Houdini, I wanted to learn all that the software had to offer. So naturally, grooming was next on the to-do list!

I had a lot of fun learning about clumping and all of the different intricacies of grooming.

Here are the final renders:

The cat model was sourced from Substance 3D Assets!