Nyan Heroes
Summary
Platform: PC (Epic)
Studio: 9 Lives Interactive
Engine: Unreal Engine 5
Released: 2024 (series of 4 pre-alpha demos)
Role: Principal Technical Game Designer
Areas: Game Modes, Core Gameplay, Systems, Prototyping
Impact: Combined technical & player-first design to help shape a quirky, competitive shooter with emotional appeal
Nyan Heroes is a team-based hero shooter where cats pilot giant mechs in the fight to reclaim their world, Nekovia.
Involved just after inception as the game's Principal Technical Game Designer since April 2022, I am mainly responsible for:
-
Designing + building all game modes
-
Parts of the core gameplay (cats!)
-
Game objects/interactibles
-
Supporting other departments with implementation and optimisation
-
Designing/Prototyping gameplay (systems)
-
Supporting other designers with Unreal/Blueprint logic
-
Writing and implementing VO lines, and (placeholder) SFX/VFX
-
Ensuring designer blueprint logic is performant
-
Representing the design team with regards to technical issues such as workflow/source control
-
Managing small team of designers
I work closely with all disciplines, collaborating especially closely with the engineering team to create a usable 'game mode toolkit' to facilitate rapid prototyping, and support elements that weren't possible/viable in blueprints only.
Charity is also a focus for the project. So far a total of $350k was donated to cat rescue efforts.
Overview
Open pre-alpha Demos
Nyan Heroes first went live on the Epic store in March 2024 for a 2 week test.
We have subsequently released 3 more demos, the most recent one just completed in April 2025.
Highlights of the competitive eSports tournament in Demo 1 (2024)
EGS Top 10 wishlisted
Since the first playtest, Nyan Heroes has consistently featured in the Epic Games Store upcoming wishlisted top 10.
Nyan Heroes does feature crypto elements, however the goal was always to appeal to regular 'web 2' gamers.
Epic Wishlist top 10 since Q2 2024. Screenshot taken in May 2025
Streamer game reactions
Compilation of streamer reactions to the first demo in 2024
Gameplay
Ejecting/Cat Gameplay
The Brief
Initially, only the mechs were playable and we started to wonder: "what if the cats could eject and were playable too?". I was asked to prototype the associated gameplay, along with the perennial design question: "will it be fun?"
And so I designed and prototyped:
-
Cat gameplay (starting off using a scaled down mech pawn with a cat mesh)
-
Locomotion, Health, Combat, Feedback
-
-
Ejecting.
-
Auto-eject if the player's mech was destroyed, but also
-
manual eject, available on [Z] if health is below x %
-
-
Re-meching. I created a gauge that fills up over time. When full, the player can summon their mech with [Z] and continue the battle.
Ejecting & re-meching sequence
The Challenge
We wanted to strike a balance between encouraging the player to get involved with the battle, but not make it such that it was overpowered and annoying to enemies. On the other hand, we wanted to retain the player's feeling of vulnerability when playing as a cat (juxtaposed against the powerful mechs), yet give the cat escape and hiding potential.
Design Solutions
A couple of systems I came up with addressed certain concerns:
1. Initially, we found that players tended to cower and hide after ejecting. To promote engagement, I implemented a system where doing damage also contributed to the re-mech gauge filling up.
Mech-riding gives invulnerability to the cat
2. Some players were missing the correlation between doing damage and the re-mech timer reducing. To help with this I proposed we added a flash to that UI element on doing damage, so the player starts associating these together.
3. We also found that players were getting increasingly cautious as their gauges were filling up, as being killed close to a re-mech was considered as unfair (essentially, experienced as additional respawn time). To alleviate this, I came up with a system whereby the player's respawn time was dynamic, and took the player's cat survival time into consideration (i.e. longer cat survival = shorter respawn).
4. To promote strategic manual ejection, players that ejected manually received full health as a cat, where players that auto-ejected only receives a fraction.
The Result
And so I also ended up being responsible for the cat gameplay overall, including camera, controls, balancing stats/abilities etc. Later I was also involved with designing the mech riding.
I'm satisfied that I've succeeded in creating systems that result in the player surviving the battle as a cat and successfully re-meching really gives a great sense of satisfaction.
Ultimately I'm proud that these systems made the game really stand out in terms of giving it more identity and a bit of a USP.
Cat locomotion
Game Objects
I was responsible for building, or at least prototyping, all of the game objects used in Nyan Heroes.
This includes things like:
-
pickups
-
interactible/moving objects
-
game objectives
-
cosmetics
-
environmental hazards
The Fishball
The Payload
Example of some cosmetic pickups I wired up in my test level
Function I wrote to handle bounce SFX and VFX, where the frequency is limtited to prevent it from going mental
Reactive hanging Balls
Game Systems
I was also responsible for designing the game's communication systems, including Pings, Emotes and the Chat system.
I designed the ping/emote system in such a way that it was flexible for designers to set up. Each element had several attributes, including:
-
Ping [MMB] or Emote [V]
-
Icon
-
Position (in game-world at trace hit or attached to player)
-
Shortcut (for example, call for Heals is also on [C])
-
Audio/VO (optional)
-
Animation (optional)
-
Chat message
-
Audience (Allies only/everyone)
My favourite emotes are "Fierce", and "Laughing",in which the player-character can be heard hissing/wailing and purring respectively.
Select "Fierce" to hiss at your opponents
Game Modes
Overview
I was responsible for building all of the game modes in Nyan Heroes. The main modes that made it to the live releases are:
-
Stronghold (Yucatán)
-
Payload (Antarctica)
-
Fishball (Neko Tokyo)
Stronghold takes place over three symmetrical maps, where each team fights to control a central objective, King of the Hill-style.
First team to two wins takes the win.
I will elaborate a bit further on the other two main modes down below.
Early Prototyping
WiP Prototype match from mid 2023. Dev playtest.
Change of Direction
Although we felt this mode had potential, it was in its current form too complicated for what we wanted, and so suffered in terms of accessibility. There were 4 teams, and multiple factors contributing to the team scores, including kills and fish deliveries.
Therefore we decided to boil things back, reduce the team count to 2 and try out modes with simpler objectives to shift the focus back to let the gun-play and movement shine while we experimented with more unique game modes in the background.
When we initially started off, we were aiming for a Battle Royale game. Then that evolved into a 4 team-based BR, before we decided to go the objective-based route.
Each step of the way I designed and wired up the game modes, created all associated game objects, created (placeholder) UI, notifications and implemented all the VOs, SFX and VFX.
I also helped to run the regular playtests, and collated, digested and acted on feedback.
An example of an early prototype is this 4 team fish extraction effort.
Payload
Overview
First introduced in pre-alpha playtest 2, Payload takes place in an arctic research facility, known as Erebus.
Inspired by the Team Fortress 2 classic of the same name, the attacking team must escort the payload forwards into the enemy base, while the defending team tries to stops it.
I set up the game mode itself, as well as all the associated game objects (the payload with its behaviour, variable speeds, health buff, checkpoint system - which itself controls the gates and unlocks secondary base, and the game rules/win conditions).
I also prototyped the UI, and balanced the experience.
The Payload
Balancing/setup example: Round Over
A lot of my day-to-day work involves balancing and polishing the game mode. Here is an example showcasing a specific sequence that I set up. The video here shows me tuning the round over sequence setup and timings:
-
Attacking Player escorting payload
-
Overtime (keep going while an attacking player is still escorting the payload)
-
Blinking hazard lights (plus SFX with falloff)
-
Explosion (camera shake, SFX, VFX + radial force)
-
Spawn diegetic marker at place of payload to show distance in Round 2
-
Round Over VO + Notification
-
Fade to black
-
Switcheroo of sides (teleport players to new bases + reset of all game objects)
-
Unfade
-
Switch phase to warmup
-
Initiate countdown to next round
Round over sequence
The corresponding round over logic can be seen in this image, in the game mode blueprint and the payload blueprint, respectively.
Round over logic
Overtime logic
Issues in Live
Initially, the mode took place over two rounds since the mode is asymmetrical and we wanted it to be as fair/competitive as possible.
However, player feedback indicated that the perception was that in some cases the matches were going on for too long. Team balancing was limited in Quickplay at the time, and so in cases where matches were a foredrawn conclusion, this mode could feel like it was dragging on.
Design Resolution
Matchmaking/balance improvements aside (something that would come with increased player numbers), I decided for now to alleviate the issue by compromising on absolute competitiveness in Quickplay (which is designed to be casual in nature). I set up the mode in such a way that for Quickplay it could be configured to function in a single round. In this case, the attacking team wins by reaching the end of the track.
Single round Payload dev test
Custom matches, used for competitive tournaments, still have the option to play 2 rounds per match, as could the upcoming ranked mode.
Fishball
Overview
Switching up the tempo compared to the other modes, Fishball is an intense cat-themed e-sport situated in downtown Neko Tokyo.
Each team must score goals using a pufferfish-shaped ball in the opposing goal. First to 5 wins (or most scored when the timer is up).
More experimental in nature than the previous two modes, this was my favourite to work on.
Carrier debuffs applied using Unreal's Gameplay Ability System
Experimental Gameplay
Dev playtest of Fishball game mode. Recorded early 2025
More experimental than the existing previous modes, I set the Fishball mode up to quickly test out various configurations. A-B testing, as it were. I exposed relevant variables to PlayFab profiles, and configured these to have different setups so they could be changed on the fly without needing new builds in between.
For example, here we can see that there are a couple of different carrier debuff settings. In the first the player cannot shoot, but can use their movement abilities, in the second the opposite is true (we went with the latter in the end!).
I wired up all the game objects such as the goals and ball in blueprints, and handled all the scoring in blueprints too, which allowed for quick prototyping/iteration.
Example of organic improvments
What I particularly liked about the mode is the sudden death feature I came up with to resolve draws, which functions in both senses of the term; respawns are disabled, so either last team standing or next team to score wins - whichever comes first.
This ended up leading to some very close and tense situations!
GOALLLL!
Misc
NFT Generator
Overview
Back in 2022, when NFTs were still hot (ahem), I was asked to create a system that would generate 11098 unique guadian setups. The variables included 3 types of attachment, series of decals 'cattoos', and a series of raries and different materials for each. Each NFT included 3 images with different poses, and so my total output was 33294 unique high resolution images.
Using only blueprint logic, I successfully generated the images, which formed the NH Genesis release, and can still be seen on sites like Magic Eden.
I approached it by importing the pre-calculated guardian setups from Excel via a data table into a special Unreal map. The blueprint script then took the instructions from each line and configured a mech in a scene as specified in the datatable, and selected an appropriate camera based on shot, mech type, pose, attachment, background, cattoos etc.
Then it took a high resolution screenshot of each and saved it with a specific descriptive name that the blockchain crew used to do their magic, before moving onto the next one.
Setup
Skin designer
Overview
As part of an experiment in early 2023, I was asked to design and prototype a custom skin creator, as a bit of a concept/demo to show what could be done in the future if we allowed players to create and customise their own skins using user generated textures.
I later created an updated version where I used Unreal's savegame feature to remember player's setups. In this video I'm skipping through some of my favourite creations.
Iteration
"Got any catnip on ye?"
Later iteration includes saved setups cat paws in UI
Project End
Unfortunately, as is seems to be the way these days, shortly after we successfully completed playtest 4 we got the unexpected news that the game was to be canned.
My understanding is that project funding simply ran out, and despite the retention numbers being solid in the last playtest, the overall player numbers were underwhelming compared to the previous tests. As such, we displeased stakeholders and failed to secure the additional investment necessary to continue.
Reading in between the lines, I can only speculate that the dropping token value, crypto market conditions and dropship farmers may have also played a role.
Either way, I think it is unfortunate that the project has been iced as it had so much potential, and we had such big plans for the game and its loyal community.
While I regret that Nyan Heroes didn't get to see a full release, I feel very priveleged to have been a part of it. Nyan Heroes was an extremely challenging and exceptionally fun project to work on with an incredibly talented team.
Final Thoughts
Learnings
Ending on a positive note, here is a summary of my takeaways from working on Nyan Heroes for 3+ years.
-
Git doesn't play nicely with Unreal. From a practical regular designer/artist PoV at least it doesn't. Usability issues can be alleviated but involves a lot of costly bespoke customisation.
-
Sometimes quick hacks end up in production (despite the intention to tidy up later). For example, creating a test item variant using branches inside the same BP can evolve into a spaghettified mess. It is often tidier and more future-proof to create a child blueprint that defines a variant's behaviour.
-
Although crypto games have a bad reputation at the moment (with some bad/exploitative examples in the wild), it is totally possible to create a decent/fun quality game in its own right that is monetised ethically using crypto components.
-
A remote studio is a viable setup. In my experience, it works best when devs have high levels of: seniority, self-motivation, flexibility, empathy and communication skills.
Final (upcoming launch teaser) trailer