Undead Depot (now titled Service with a Shotgun) is a visual novel combined with an on-rails shooter.

It was also my first ever paid experience in game development!

I wanted to make a visual novel that anybody could get behind. A lot of my friends didn’t like visual novels because they were too text-heavy, on top of having no proper gameplay loop to keep them engaged.

It came to me while I was working at a gas station, and realizing the fun of multi-tasking customers and managing the store. Running back and forth while still getting a few conversations in with some of the regulars I liked, hearing updates on their lives, it was a lot of fun. “Bingo! That’s it!” The next day I cobbled together a prototype and Service with a Shotgun was born.

Originally starting out as a failed student assignment in late 2023, I stuck with it and eventually finished a demo for Kickstarter in early 2024. The kickstarter failed, but a then small, new publishing company, Silver Lining Interactive, liked what they saw and offered to publish Undead Depot! I offered to develop 5 chapters for the game, with each chapter having a different location and gameplay mechanics to switch things up. They gave me the go-ahead and I got to work.

Once I started I didn’t stop, giving 100% into this game (now titled Service with a Shotgun) until it was released in November of 2025.

A major philosophy behind the development of this game (and my games in general) is thorough refinement and plentiful playtesting. I would test the writing, gameplay, conveyance, and many other parts of the game to make sure the people were getting a finely tuned experience.

One of the biggest hurdles I had while developing Service with a Shotgun was overwhelming the player. Zombies would give money at random and it was very easy to run out of ammo and get completely overrun.

My fix was to change the money distribution (5 for body 10 for headshots), and have the zombies spawn more dynamically. The spawning worked as a rubber-band. The more you killed, the less they spawned, eventually rubber-banding back to the default spawn rate after a while.

Genre: Visual Novel/First Person Shooter

Team Size: 6

Roles: Sole Designer/Developer/Writer

Time: 2 years

Engine: Unreal Engine 5.2

Grind is an arcade skater inspired by the PS1-era of Tony Hawk games.

I had about 1 month to make this game, and it had to be great.

The game has about 6 total unique moves:

1. Grinding

2. Speed Boosts through landing 180s

3. Wallrunning

4. Wall Climbing

5. Bullet Time

6. Double Jumping through Bullet Time

I wanted to make a very diverse and easy to pick up but hard to master moveset, to do this I gave every move its own extra sub-ability (like double jumping through bullet time) and techs that could be combo’d into each other to make lines throughout the level.

A majority of time was spent on nailing the movement, with two weeks put into that, one week put into developing and refining a blockmesh, and an extra week put into polish and visuals.

One problem that kept popping up was the game being too fast for some people. My solution to this ended up being the Bullet Time mechanic, players could slow down time midair or on a rail, this also worked to let them redirect themselves either on a rail by switching directions or midair with the double jump.

A lot of time went into the level and making sure that all of the rails, shortcuts, walls and general layout naturally flowed, allowing for long combos and plenty of routes allowing the player freedom in how they tackled their lines.

It was important to not let feature creep keep this game from ending up bloated and un-polished, instead giving it time to be the best little proof of concept it could be.

The last week of the game went into polish, where I would work tirelessly to get it as shiny as could be.

I added a main menu, loading screen, settings with camera/acceleration options, tiny details into the level to make it feel dirty and lived in, polished general movement, special effects, and revising controller support.

Genre: Sports/Arcade

Team Size: 1

Roles: Sole creator

Time: 4 weeks

Engine: Unreal Engine 5.2

Frag Out is a movement/wave shooter where your enemies can all explode into each other, creating massive combo kills.

The biggest challenge was obviously time constraints. For this reason, I had to speedrun a lot of the planning process, making this hilarious game design document.

I would work on getting the movement feeling as good as possible. While doing that, I would work on getting the enemies to spawn properly.

Three problems I kept running into throughout all the playtests I would do:

1. Players not knowing where the enemies were (and them sneaking up on players)

-My solution was to make the music dynamically shift
depending on how close the enemies were and what
direction they were coming from (imagine the nearest
enemy to the player is holding a speaker with the music).

2. Players camping in spots, putting the enemies in a bottleneck.

-For the players camping in spots, I made the enemies
spawn in 3 out of 8 possible spawn points at random to
keep players on their toes.

3. Players wouldn’t rocket jump with the shotgun.

-With players not rocket jumping, I spawned the players in
a tight space, and a crack on the floor. Instead of the crack
breaking when shot, players would rocket jump due to the
tight space.

A week later, I had finished the blockmesh, did lots of refining, and then I had about one more week to add polish. During that week I added:

1. Controller support with aim assist.

-The aim assist was the last thing I added, after

2. Varied locations throughout the small level for better
navigation.

3. Loading screen with custom designs.

4. Extra optimization reaching speeds of 100fps.

Genre: First Person Shooter

Team Size: 1

Roles: Sole creator

Time: 4 weeks

Engine: Unreal Engine 5.2

Don’t Drown is a small isometric puzzle game with the premise of (as the name suggests) not drowning.

I wanted to make a simple concept explored thoroughly. In this case, the concept was navigating underwater spaces with limited oxygen. Each move you make underwater removes an oxygen bubble, only refilling through bubbles you find or by resurfacing.

A problem that continued in the playtests were player's being confused on how much oxygen they had left.

To combat this, I made 3 extra levels exclusively for giving the player time to realize moving took away oxygen as well as a signifier to alert the player when they begin to run low on oxygen.

I wanted to pace the game in the same vein as Portal 1’s progression.

1. Introduce an idea.

2. Let the player mess around with the idea.

3. Have player master the idea.

4. Combine the idea with previously established mechanics.

On top of introducing the idea, I wanted to tell a little story through the limited environment: a diver goes inside of an ancient temple looking for treasure.

This narrative would make the introductions of the mechanics feel less like a learning experience and more like a natural flow. You aren’t playing a puzzle game, you’re searching for long lost treasure!

Focusing on short and sweet taught me a valuable lesson about polish: There’s never enough. With a week left after developing all the levels and mechanics; I decided to focus that time on refining the mechanics and game feel of this little game.

To add polish, I developed:

1. A HUD that responded to the player moving in and out
of water with bubbles rising, a blue filter, and a
splashing sound effect.

2. Moving tiles that would raise upon diving underwater,
potentially revealing secret routes to the player.

3. Dynamic music (courtesy of Joey Azevedo) that would
transition to and from an underwater variant
depending on what state the player switched to.

4. Sunken treasure at the end, of course!

Genre: Puzzle

Team Size: 1

Roles: Sole creator

Time: 4 weeks

Engine: Unreal Engine 5.2