LEVEL DESIGN

My approach to level design changes depending on the game and the specific situation.

My general level design workflow:

Step 1: Define the narrative context

Is the game set in a cramped cave or an open field?
Is the intent to scare the player, or make them feel powerful?
What comes before and after this section?

The narrative context shapes the experience the level should deliver.

Step 2: Identify the game mechanics

What abilities does the player have at this point?
What level design bricks are specific to this environment?
Are there mechanics that have been overused or underused elsewhere?

Thinking about mechanics early helps create a coherent and memorable environment.

Step 3: Generate interesting situations

With the context and mechanics defined, I brainstorm situations with as few constraints as possible. I usually start with pen and paper for speed, then move to Illustrator or Figma for more complex layouts, or directly into the engine for 3D ideas. This step is also key for sharing ideas with the rest of the team. Early feedback from artists, programmers, and other designers helps refine concepts and keep the project cohesive.

Step 4: Build the gameflow

At this stage, I balance multiple factors:

Introducing new mechanics progressively.
Alternating pacing between discovery, mastery, and rest.
Ensuring environmental storytelling aligns with the narrative.

I select the most fitting ideas from Step 3 and arrange them into an initial level schematic.

Step 5: Assemble the level

I greybox the level with the intentions, gameflow, and narrative in mind, guiding the player through landmarks, geometry, and visual cues. Communication with artists and narrative designers remains constant during this stage.

Step 6: Playtest and iterate

As soon as a playable section exists, I put it in front of players to identify issues early. I then repeat Steps 5 and 6 until the level is both functional and polished, gradually replacing placeholders with final assets.

My creative process for Twin Cores:

I chose to use my graduation project Twin Cores as an example, as I feel this was a major step forward in my growth as a level designer.

I needed to create an environment that challenged both the player controlling the ship and the player controlling the weapon. That meant including both hazards to dodge and targets to shoot. The “environmental corruption” that players were fighting also had to feel more present here than in other areas. I decided to set the level in deep ravines rather than the open skies seen earlier. The close walls and ground enhanced the sense of speed, provided a distinct visual identity, and supported narrative moments with enemies jumping in and out of cover.

I used the ravines to provide cover for enemies and create obstacles to dodge. I also featured the snake-like enemy type, representing heavily corrupted foes. This enemy was made with this section in mind, and was purposfully rare in other areas of the game so as to give it a strong association with this environment. Its long body fit naturally into the ravine corridors, and its attack patterns worked well with the fused-ship mechanic.

I imagined sequences with enemies ambushing from the ravines, fast-paced turns, and weaving between rocks and incoming fire. These ideas became the basis for the cinematic feel of the level.

Balancing the introduction of the snake enemy, close-quarters ravines, and the fused-ship mechanic was a delicate task. I built several versions of the gameflow, mixing and matching sequences until I was happy, then refined it further through playtesting.

One challenge was our professors’ warning about rail-shooters, as previous student rail-shooter projects had struggled with “empty” environments. To address this, we created a looping environment instead of a straight path, reducing unused space and increasing density. My greybox had players first flying inside the ravines, then looping back above them for the return.

I noticed during playtesting an issue where turrets at the end of a ravine fired directly toward the players. The weapon player’s targeting cursor often blocked these projectiles from the pilot’s view, making them nearly impossible to dodge. I solved this by moving the enemies above the ravine so their shots crossed the screen visibly, and by spreading them out along the path, making them easier and more satisfying to target.