top of page



This 3rd person adventure level design brings players to the heart of a Favela. Where our player needs to escape the island as the enemies are looking for the artefact they stole.

Navigate a web of narrow alleys and staircases, offering multiple routes for stealthy exploration. Every decision shapes your journey and branches you to new possibilities, providing a dynamic and immersive experience.

Time indicator




Unreal logo




I started with designing a high level network that focused on points of interests. After the network I worked it out with giving the points of interests shape. From there I had the first sketch done. I would go into engine to figure out the scaling and go back to the drawing board to iterate on the sketches. The iterations were mainly focused on the flow that I could create through vertical spaces.

The initial (left) and the final tension (right) graph divided by gameplay beats

(move the middle bar with mouse to see the differences)


The first big change was the first enemy introduction. The onboarding for the player was too hard and had to make this interaction more easy for the player to do. I changed the traversal to be more linear and the AI in a static position to make the onboarding experience more effective.


The second big change in the graph was focused on the reaction of the players as they felt way more tensed than expected which I thought was a positive change for the tension graph and embraced the change.


Lastly, players were able to finish the last beat without much sweat so I added more enemy patrol sections to investigate in player's area to make that last moment really tense.



The landmark gets introduced in the first minute of the gameplay and will reappear with each level beat to help players understand the space around them and give them a sense of progression as they get closer to their goal.

I also used the landmark as a vista when I wanted to slow down the player and take away the intensity.


Multiple techniques have been used in the level for guiding the player.


  • Guiding lines visualized by electric cords as a subtle push. These will act as action lines that go hand in hand with the meshes that create flow.

  • Enemies when I need to guide the player more directly to places they might not tend to go such as going up when they've been going down for a while.


After the navigation beats, I transitioned the beat into an defend/attack beat. Creating a new biome that looked vastly different would help communicate progression to the player. I stepped away from the tight spaces and created a market space that had his own identity that still fits the Favela-feel. This way the player knows they are closer to their goal but also can expect new challenges.


To give the player some more agency over their experience, I created branching paths.


From the first encounter, the player can decide where to go. Both will have the same amount of enemies and no new mechanics introduced but the player learns the structure of the space better and learns how different paths connect each other. This prepares the player for upcoming choices where multiple paths need to be explored.



The first combat encounter introduces the player in showing who is the enemy through an event where you see/hear them threatening a civilian.


The player then has a learning opportunity to stealthily takedown the enemy which will set the tone for the rest of the level.


The player is being put in the last tenseful event of the level where they have to wait out the waves of enemies in a more open area until they get picked up by boat. 


This design choice was made to showcase the previous skills that were learned, give emotional engagement and narrative coherence to create a memorable conclusion to the level.

Breakdown of courtyard


This being the second interaction, I still wanted to give the player the advantage by giving them windows to predict the enemies movement as well as a vantage point that would accommodate the rusher and stealth playstyles.

The riskier path makes you more visible to the enemy but gets you faster through the area while the other path is slower and you have to strategize your plan more. With the reward of making the next enemy interaction that's in the next room easier as you can take more time to plan that out.

Breakdown of market


Starting from a vantage point you have two route possibilities. Depending on which branching path you took in the previous section you will either be pushed to the left or right direction through flow.

The first route is easy to go in as it is not blocked by any challenges but from it's exit it will be harder to navigate through cause at this point an alarm has gone off and reinforcements have occupied the space.

The second route has a flanking area that helps the player find for different ways into the building but then getting to the second entrance gets harder as they have to cross the market diagonally. 

Breakdown of side route


The player has had, beforehand, a lot of exploration, platforming challenges and a narrative beat. So to mix that up I gave the player an option to continue with the same playstyle that I offered or give an option where players could rush in and get the satisfaction of the kill. 

The takedown opportunities are reachable through multiple routes. This way the player can circle the crates to make sure they are not caught but at the same time still perform their action.

Breakdown of harbor


To have a climatic ending. I offered a playable space where the players could test their combat skills that they have picked up throughout the level. To this point they should know how the enemy behaves and how the spaces create loops for flow to shake off enemies.

From there I created a surprise by raising the tension with having waves of enemies coming in with no exit for the player. They will need to wait out the waves in order to be picked up by boat and ending the level with a nice scenery of the whole level.




I divided the sections based on a narrative structure with a beginning, middle and end. Once I started scaling the level, I had to rescope as the challenges would be bigger than the sketch sections I made with planes in engine. Once I had a better sense of the structure in the first part, I would get faster in the second. The iterations in the first part were mostly focused on the challenges, making sure the player wouldn't feel it too repetitive.


Most iterations in the second section (market area) were focused on enemy placement as well as flow throughout the level towards the goal (the keys). Some extra guiding was needed as a feedback result from playtesting which I did by making the electric cords connect to the goals. This way, the player wouldn't lose their agency as they could still decide where to go based on the choice that the electric cords gave.

Progression on level from scaling to end result


The biggest iteration on this part of the level was the amount of branching paths that I designed. I initially designed three but ended up with two to reduce scope

Another issue with the first version was that in one of the branching paths there was a new mechanic that the player gets onboarded on. But if the player would choose another path, they would have never learned this mechanic. With the newer iteration, I focused on making sure that both paths have the same learnings and only give the player a choice in what type of gameplay they would like to get out of the level. A more combat route or more navigation route.

Iterations on the first area focused on branching


Using the paint overs, I made a modular kit with metrics based on the houses of Favela with the Unreal Engine modelling tool. 


After that, I made some extra assets with the focus on purpose. Such as market stalls and crates as covers. Then creating some secondary assets to enhance the favela aesthetic such as tv antennas, control board and a boat.

The foliage is from an asset pack by craftpix on

Modular kit used top view
Modular kit used side view



To communicate to the player that they are stuck here, there is a scripted event that closes the big door. The player can see beforehand the open doors through the windows and have to find a different way out.


Once the player finds a way in, they override the electric system. An alarm goes off with an announcer mentioning that they require backup. An electric door opens followed by enemies coming in. This trigger will also impact the market area as it will spawn more enemies from this moment. This scripted event was added to add some complexity and foreshadowing their way back to the market area.


To add some foreshadowing for the harbor challenge where the player has to wait out enemy waves until a boat picks them up, the player gets informed by their friendly NPC. They mention it will take longer for them to pick them up and that they have to hold on.


Gathering references

For gathering references, I would delve myself into google maps street view to get a sense of the space and structure of the place. I also used it to understand the relationship between land and the sea that I would be able to take advantage of.

Paint overs

To understand the housing in the area and imagine myself climbing the buildings, I made a few paint overs to get an idea of the ledges and window placement to make the experience more believable.

Reference images



What I would have done differently next time is to design encounter sketches before designing the layout. I focused a lot on the flow of the level which would often be contradicting with the enemy's design. This made me iterate a lot more on the spaces that could have been reduced.

My biggest learning is about giving enough space for different player styles to reanalyze their decisions and make their choice feel more impactful. I also learned about the benefits of scoping down early in the project but still maintaining the same gameplay experience.

bottom of page