Tennessee Luc Jackson Phillips Ward

Game Designer
Technical Specialization


Professional Projects

Minecraft Dungeons 2
Approachable ARPG

• Shipped Sept. 29, 2026
• Roden Games in partnership with Mojang
• 01/2025 - Present

My Role:
• Game Designer

Design Goals
• Create novel enemies that expand on the suite from Dungeons 1, including those from Minecraft’s newest dimension.
• Upgrade boss encounters to an engaging and challenging status.
• Balance enemies and bosses for all player counts (1-4).

Learning Outcomes
• How to pace and maintain engagement during high level encounters.
• Design philosophies for making individual enemy units click together.
• Integrating and facilitating collaboration between remote and in-person teams.


Additional Projects

Marked for Deletion (2024)
Meta Stealth Game

My Role:
• Technical Game Designer

Design Goals
• Define game development into ‘breakable’ sections.
• Allow player freedom in the order they eliminate targets.
• Make each elimination's effect visually obvious and meaningfully increase the challenge of future eliminations.

Learning Outcomes
• How to chunk complexity of the game director’s creative vision.
• Emphasizing novelty through content over mechanical depth.
• Fundamental properties of stealth games and how to set an experience on a scale from predator to prey.


Murder on the Rocks (2024)
Bartending Murder Mystery Game

My Roles:
• Lead Game Designer
• Technical Game Designer

Design Goals
• Interweave gameplay and narrative so that each influences the other.
• Create tactile feedback for slower paced gameplay elements.

Learning Outcomes
• Concepting experiences with limited resources in mind.
• Maintaining core elements of features when downscoping.
• How to rely on the strengths of the team and emphasize those aspects.


Down the Hill (2024)
Mobile Racing Game

My Roles:
• Game Designer
• Level Generation Programmer

Design Goals
• Create an experience emblematic of a real life sport - soap box racing.
• Approach design and development from a mobile first mindset.
• Find the fun in a narrow set of restrictions.

Learning Outcomes
• How to iterate quickly when pulling from unfamiliar references.
• Be more open about suggestions, especially when I’m confident in my ability to implement them.


Hunting the Aurora (2021)
Top Down Action Game

My Roles:
• Game Director
• Lead Game Designer

Design Goals
• Define varied enemy types that can be implemented in parallel.
• Create readable encounters using the various enemies as puzzle pieces.
• Ramp complexity of combat to culminate in a boss encounter that builds on prior player knowledge.

Learning Outcomes
• Using environmental elements to make existing content feel fresh.
• How to define a minimum viable product with remaining project time in mind.
• Drive player understanding of mechanics through audio and visual feedback.


About Me

     Hailing from California, I've moved to Sweden to study and work as a game designer.
     I break down games into their base elements, analyze, and reassemble to create new mechanics, systems, and memorable content. I'm no Seagull, though! I'm always in engine tinkering to make my designs feel as great as possible.
     I am energized by working with great minds across all disciplines, and at all team sizes. I love the role of designer because it positions me where I can collaborate with everyone in creative ways.
     Currently I'm working with Roden Games and Mojang as a Game Designer on Minecraft: Dungeons 2. Enemy design and implementation is the bulk of my role, and I also routinely assist with the broader game systems.

Minecraft Dungeons 2 Trailer

Minecraft Dungeons 2
Approachable ARPG

Overview

Dungeons 2 is a sequel to Minecraft Dungeons, an approachable ARPG. It expands upon the first game in several notable ways including an overworld, Soul Corrupted enemies, and new gear and equipment slots.

Details

Shipped Sept. 29th, 2026
DLC 1 and 2 coming soon!

My Role as Designer

When first joining the project, I worked to design and implement various enemies and combat encounters, focusing on single player balance.After gaining more familiarity, I shifted focus to many high level encounters including boss fights and other difficult enemies. Part of this focus is to maintain the difficulty in single player while updating it in multiplayer such that the experiences roughly match each other.I am currently leading the design and implementation of some of our post launch content - more details to be revealed soon!


Design Process

Overview

Throughout the course of this project, I have contributed to all stages of enemy design and creation. Some of the highlights include the spawning system for Soul Corrupted enemies, boss designs, and multiplayer focused balancing.

Soul Corrupted Enemies

Soul Corrupted Enemies are a new addition to the series and act as elite enemy types. Every normal and mini-boss enemy has a soul corrupted version which increases several aspects of their challenge. They are empowered by stat bonuses as well as a new, randomized ability.When I joined the project, the core mechanic of soul corrupted enemies was defined, and implementation was in progress. I worked to balance and re-tune each of the soul corrupted enemies, including their spawn frequency, spawn counts, stat bonuses, and new abilities. While balancing these enemies, I noticed that not all of the randomized abilities functioned on every enemy type. For example, some of the abilities only triggered when the enemy used a melee attack, but many enemies only had ranged attacks or were more focused on utility. At first I worked to minimize the negative impact of mismatched abilities and enemies types. Since melee and ranged enemies made up the bulk of our collection, I updated the existing abilities to trigger on both types of attacks.I then prototyped a whitelist for enemies that would only allow usable abilities to be applied to them. The Creeper, for example, was set up in such a way that its explosion did not count as a melee attack, and therefore should only gain abilities that did not require one to trigger. We had designed several of these abilities to be universal, so any mob could utilize their effects. With the help of our engineers, we fleshed out that prototype into the finalized whitelist for spawning these soul corrupted enemies, and could continue to their final balance pass.

Boss Designs

Similar to the Soul Corrupted enemies, most of the bosses for the base game of Dungeons 2 were fully functioning when I joined. However, none had been balanced specifically with multiplayer in mind, and several had significant pacing and flow issues. We were also rapidly iterating on our leveling and scaling systems, so my approach to balancing had three major phases: pacing, level scaling, and player count scaling.The first phase was to smooth out some of the pacing and flow inconsistencies. Some bosses would repeat certain abilities too often, or rarely use an ability at all during their encounter. Much of this work involved modifying the boss’s state tree, or editing the cooldowns of abilities. Additionally, I tried to determine if any abilities should have a distance requirement. Some ranged abilities, for example, I prevented from activating if the target was too close to the boss. Once I felt confident the pacing was improved, I looked for external confirmation. Ideally, I would be able to source feedback from people outside the Mobs team, but I quickly learned that general callouts for feedback didn’t get much of a response. Testing within my team would be sufficient for the time being, but I started to understand that, in order to get more focused feedback, I would need to at least reach out to other teams.

Challenge: Siloed Teams

After reworking the ability cadence for the bosses, I turned my attention to their balance in regards to our leveling systems. Because these systems were actively being worked on, I needed to work closely with our other design teams to understand and keep up to date with the current implementation of our level scaling. The structure of teams working on Dungeons 2 was, unfortunately, fairly siloed overall. While I don’t think I have enough perspective on the project to speculate as to why, I did notice gaps in communication and ownership between teams. I was, of course, not the only one to recognize this issue. Multiple team restructures were attempted to try to alleviate these gaps, but, from my perspective, did not resolve them.One of the consequences of this siloing was a lack of support for more comprehensive testing methods. We had the ability to set up test arenas for bosses, and could create progression commands, but neither of these gave an accurate representation of player gear or level when they would encounter these bosses. I often used the test arenas to exclusively tune the flow of abilities as mentioned above, but could not use them to effectively balance level scaling. While the test arenas were all uniform in shape and size, the progression commands helped to playtest bosses in their more unique, personalized arenas. The progression commands also completed any requisite quests and gave some XP and gear to test the encounter with. The major issue was the lack of granularity with these commands. While we could set the resulting level for a given command, we could not choose the selection of gear. Because gear selection in Dungeons 2 can be the difference between being steamrolled by a boss and killing it in one hit, this lack of granularity made it difficult to assess the relative difficulty an average player might have with any given boss.After the team restructures, it was unclear which team actually owned these progression commands, and so they became neglected. I tried to maintain parity whenever adding a new boss encounter so that each at least had a progression command, but I was unable to dive into the system enough to make major updates. In the end, we tried to rely on our QA’s complete playthroughs to report some data on expected player level and gear selection when encountering each boss. I updated the progression commands with the data from QA along with any playtesting data I could source. These complete playthroughs took too much time to do regularly, however, so they were infrequent. Continued into our current DLC production, our testing on this front remains largely the same, and is therefore suboptimal for balancing our mobs.

Multiplayer Balance

The next phase of my comprehensive boss balancing work focused on how we scaled combat difficulty based on the number of players in the lobby. In Dungeons 2, 1-4 players can play together in any mix of couch coop and online sessions. When looking specifically at how to modify the difficulty, we had several knobs we could edit per player count. We had some control over the number of mobs spawned (but not in every case), and we could multiply the overall values like damage, health, poise (how much damage needed to stagger a mob), etc. For most of our enemies, we stuck to modifying these three values. Because our Areas team owned mob spawning (including placement, enemy type, and enemy count), my role in this endeavor was mainly to balance the overall multiplying of base values.The way our inheritance was set up, I could apply a baseline value to groups of mobs, then make changes to the scaling values of individual enemies if needed. I used this grouping to give each class of our enemies (basic, mini-boss, boss) a different scaling value. Since we wanted to spawn more basic mobs with more players, I elected to give them the lowest increase in stats per player. Conversely, we don’t tend to spawn multiple bosses at the same time, so those were given the largest scaling values per player. This process, again, required testing with other teams to assess if our intended outcome was achieved. Since a large portion of external feedback was driven by multiplayer testing, however, I could rely more on general feedback from everyone on the project to guide some of the individual tuning. Most of the basic enemies and mini-bosses do not have individual scaling, but many of the bosses do. I was able to source more feedback on the individual difficulty of each boss encounter, and used that to make more personalized tweaks to each.

DLC 1: The Humblers Awaken

The Humblers Awaken Trailer

More information on my design work for this DLC coming soon!

Marked for Deletion (2024)
Meta Stealth Game

Overview

• Marked for Deletion is a 4th wall breaking stealth game where game developers are your targets.
• You’ve escaped out of your game and into the developers office through a forgotten bug, and now want revenge.
• As you delete each of the developers, you’ll notice that their work is deleted from the game you are actively playing!

Details

• Unreal
• 5 Weeks
• 10 People
• Windows
• Perforce

My Role as Technical Designer

• Designed and helped implement enemy AI behaviors.
• Designed and helped implement stealth gameplay loop using Unreal's Blueprints.
• Implemented some of the Developer deletion effects.

Murder on the Rocks (2024)
Bartending Murder Mystery

Overview

• You’ve found a note that one of your regular patrons is about to be murdered, and it’s your intent to stop it.
• Everyone has a motive, and it’s up to you to decide who is capable of such a deed.
• The better your drinks, the more information you’ll get from the patrons.

Details

• Unreal
• 7 Weeks
• 13 People
• Windows
• Perforce

My Role as Technical Design Lead

• Lead team vision for project and maintained team alignment.
• Designed and helped implement drink making system.
• Assisted programmers via peer programming.


Process Summary

Team Vision

Going into this game project, I had only worked with a couple of the team members previously. I knew we could rely on the narrative and art abilities of the team, so I wanted to suggest ideas that could play on those strengths. One of the possible themes was ‘Put a Twist on It,’ so I thought of making a detective game where the suspects come to the detective rather than them having to search for them. This idea led me to a bar setting, where the patrons were all possibly involved in a murder mystery.Once we had decided on a game idea, we took some time to research and play through other games which focused on narrative and had drink making, or crafting components. Red Strings Club was one of the largest inspirations for the vibe and atmosphere of the environment. I also took notes from Coffee Talk and Tavern Talk for how to build the drink crafting system.

Drink Making System

We had initially tried to create a more ambitious crafting system, which focused on tactile elements and would require manipulation of objects in 3D space. The programmers and I did have early success creating a recipe system, which kept track of the ingredients added and checked them against a table of possible drinks, but the 3D manipulation had issues. The ingredients were tricky to manipulate into place, and didn’t always stay within the bounds of the bar. As a team, we ultimately felt we would not have enough time to both implement and iterate on the system. Instead, we shifted to a UI based approach, which allowed us to keep the underlying recipe system and main loop of actions for the player.

Key Takeaways

Some of my takeaways for the project are that, while the scope was not overly ambitious, it still should have been scaled down to allow for more time for iteration. We also should have more clearly defined the ownership of our pipelines, as not having this led to friction on multiple occasions. Lastly, we were lucky not to have many troubles with bottlenecks. One example of a bottleneck was that our narrative script and implementation was entirely done by one person, and the project could have been majorly disrupted if they became unavailable.

Down the Hill (2024)
Mobile Racing Game

Overview

• Navigate the winding paths, but don’t let your car break down!
• Race through randomly generated levels that change every day.
• Fight for top spots on the leaderboard! Challenge your friends with the level seed and compare your times.

Details

• Unity
• 4 Weeks
• 16 People
• Android
• Perforce

My Role as Technical Designer

• Lead design of the project within genre and platform constraints.
• Programmed and implemented level generation.
• Designed multiplayer code sharing based on generated levels.
• Assisted in level design by creating chunks of levels.


Design Constraints

Our major design constraints for this project were that it needed to be multiplayer, playable on mobile, and be related to action sports. Once our team had decided on making a soapbox game, I started to ideate on how we could implement the multiplayer component. Our team initially had the idea of making a rally-style component where one player would drive, and the other would make callouts based on a map of the level. I had a couple concerns with this method, first among which being that map player would have little to do in between making callouts. I also worried that the driver could likely learn the levels without the need for dedicated callouts.

Level Generator and Multiplayer

In order to alleviate some of these concerns, I started working on a method to procedurally generate levels out of premade chunks. I had hoped that by randomizing the maps, we could not only prevent drivers from memorizing maps, but also encourage more communication from the map player.During our greenlight session, we were told to rework our multiplayer format so the team started brainstorming. As I had been working to define how level generation would work, I suggested utilizing this mechanic more heavily by making it the core of the multiplayer format. Instead of having one player making callouts, we could have any number of players all playing on a randomly generated, but, importantly, seeded level. This way, all players could play independently of one another, while sharing their completion times for the same level.

Soap Box Design and Feel

After working on the level generation and multiplayer components, I switched over to work on the feel of the soapbox itself. One of our programmers made a robust system for simulating the center of mass for each model of the cars, and I tweaked the ratios of this system to make sure each car had a unique, but playable feel to them. We ended up breaking away from realism in the simulation in order to have a good baseline feel, then worked together to make sure no matter which combination of parts you applied to the car, you’d be able to drive it down to the finish line.

Hunting the Aurora (2021)
Norse Inspired Action Game

Overview

• Hunting the Aurora is a top down, action game set closely following a mysterious winter storm.
• Fight through various enemies, like a hammer wielding giant and a lava eel.
• Gain skills by defeating enemies to grow your strength and unlock new combat styles.

Details

• Unity
• 9 Months
• 11 People
• Windows
• Github Desktop

My Role as Design Lead

• Led design of the project by ideating and iterating core systems.
• Worked with programmers to develop combat systems and enemy designs.
• Designed and implemented each level using tile maps and procedural spawning system.


Prototyping

For the first couple months of working together, my team created several prototypes to test various ideas and mechanics. We weren’t exactly sure what we wanted to make yet, and explored various styles of gameplay. One of the prototypes that I made during this time focused on testing a 3D game environment as well as Unity’s baked in animation tools. At this point, my team was interested in including survival elements, so this prototype featured bars for hunger and thirst, as well as a simple enemy.

I added pickups for the player to collect that would restore the draining thirst and hunger bars and an attack to ward off the enemies. I was the sole developer on this prototype and the design was a collaborative effort. Ultimately, our team felt less comfortable with 3D development, and we confirmed among ourselves that we would develop a 2D game instead.

Game Pitch Meetings

As the game was our senior group project, the initial design concepts were collaborative. After prototyping, we had a more general sense of the kind of game we wanted to make - a 2D action game - a few of us started to put together more concrete proposals of games. We pitched these concepts to each other and then voted on which project to pursue for the remainder of the school year.My first concept for Hunting the Aurora focused on exploration and gathering resources by defeating enemies.

While my pitch included ideas for various other gameplay systems, the main focus was on the map structure and creatures that inhabited the map.

The elements of the regional creatures also factored into the overall progression. Weapons obtained from one region would be strong against one neighboring region's creatures, but weak against the other. This would incentivize the player to progress through the map in a circle, but would give them the freedom to choose a region to start in, and explore at their own pace.

Design Process

After voting on the different pitches, our group decided to pursue my pitch idea, and I took on the role of Lead Designer. Because I was also the producer on this project, I wasn’t able to continue working on programming tasks like prototyping or scripting. As designer, I had a couple goals in mind for the player experience. First, I wanted the player to think about the enemies that they were fighting and engage with them in different ways. Another goal was to have the player build on their knowledge of the game and to steadily increase the complexity.The main way of having the player think about how to engage with the different enemies was their elemental types. The elements would act as a rock-paper-scissors type of system, so I wanted the player to think about which order to challenge each region. Additionally, I wanted each of the elements to have their own mechanical type of interaction with the player’s character. For example, fire enemies could cause damage over time and ice enemies could slow the player. The weapons obtained from a region would also be able to apply this interaction to enemies from the other regions. One of the ways I intended to increase complexity in this design was with the bosses.

One of the major obstacles our team faced was time constraints. We were ambitious with what we wanted to accomplish and this led to some major downsizing throughout the project’s lifecycle. While at first we thought three regions was achievable, our artists in particular had difficulty creating characters quickly, and some struggled with animation. Our team decided that we would have to continue with only one region, which meant changing major gameplay features. Rather than being able to obtain new weapons, we created a skill tree that the player could level up. I came up with most of the ideas for the unlockable perks on the skill tree, but my team also helped brainstorm, and the programmers specifically informed me on what they thought could be done quickly.

Without three regions, I felt that player engagement with the enemies would suffer. To combat this, I tried to focus instead on how the complexity of the encounters increased throughout the game. Instead of having an open map that the player could explore, I made three levels to traverse through. Our team was able to create four enemies and a boss, so later levels introduce new enemies and environmental hazards. Additionally, I wanted each of the enemies to interact with the player in a different way, and tried to have their behaviors reflect that goal.In the first region, the player mainly faces off against two enemies: one giant wielding a hammer and the other a sword and shield. These giants had the same body, so I wanted to make sure they behave differently, and had a different silhouette. Walking with his weapon across his frame, the hammer giant smashes his hammer into the ground, causing a shockwave that damages and knocks the player back. The sword giant holds both a sword and shield in front of himself, and strikes more quickly, burning the player if their sword lands a hit. In addition to having different attacks, the hammer giant also walks slower, and keeps more distance from the player when possible. The sword giant, conversely, moves in closer to the player, and circles them.

Toward the end of the first level the supporting ‘Imp’ enemy is introduced. The imp is the only ranged enemy in the game and is able to fire both red bolts at the player and fire green bolts towards other enemies to heal them. Because it is a ranged enemy and tries to support its allies, the imp will run away from the player. The last non-boss enemy is a lava eel. I wanted this enemy to act as more of a low health swarming enemy that would rush toward the player. In testing, this type of behavior didn’t feel very unique with how the player would interact with this enemy in comparison to the rest, so I also had the eel leave behind a damaging trail of lava. This change complemented the additional environmental hazards added to the second and third levels that created varied arenas to fight in.

True to the original game design document, the boss builds on the two giants’ attacks by wielding both a hammer and a sword. The boss has three total types of attacks he can perform; two of which are variations of the sword giant and hammer giant’s attacks. The sword creates a flaming slash that moves down toward the player, while the hammer creates a cone of fireballs. During the third attack, the boss slams both of his weapons together to create a ring of fire that emanates outward.

Key Takeaway

Hunting the Aurora was an interesting challenge to design. It is the largest game I have worked on, and I learned a lot about collaborating in larger groups. Though there were many changes from my initial designs for the game, I enjoyed puzzling out how to balance the time my team had with how much we could add to the game. Cutting two of the initial regions forced me to shift my design philosophies, and I think this has made me a more flexible designer.