I'm a systems designer. Progression, economy, online.

1.1What I owned

Core progression for the final year of development: XP and player level, the Techniques skill system and Adaptations. That meant deciding what Logan unlocks and when, with the combat, UI and mission design teams, so that new abilities land alongside the fights and story beats that use them. I also set suit costs and their unlock timing, brought material crates into the progression economy and designed the trophy set.

1.2Playtests

I worked with Sony's UX Research team on the questions and dashboards for each playtest, then used the results to tune progression. A playtest can only answer the questions it asks, so writing those questions was as much a part of the design work as the tuning that followed.

1.3XP and guardrails

Wolverine's fights are fixed set pieces, and enemies do not scale with the player. A player's level decides how many skill points they have, and so how many Techniques they bring to each fight. XP came from two places: the enemies every player has to defeat to move on, and XP caches hand-placed around each level to reward exploring. The more XP the caches held, the more exploring paid, and the further behind a player who ignored them would fall. I built the formula that set an XP guardrail at every checkpoint. It had to sit above players who found every XP cache, so it never took what they earned, and low enough that a player who died on purpose to farm respawned enemies could not overlevel.

2.1What I worked on

I joined as senior game designer on the online and UI systems: co-op and PvP multiplayer, crossplay and the UI framework behind them. I worked on the most complex menus with the UI team, including inventory management and the front-end flow. I also helped set up a review of UI screens before they were built, which caught structural problems while they were still cheap to fix.

2.2The tutorial

At launch the tutorial took about an hour to complete. Nearly every system in Legends was new to players coming from Minecraft, and the concern that they would not understand them had kept the tutorial growing. The audience skewed young, and long tutorial text often went unread. Minecraft itself teaches very little up front and leaves players to learn by playing. After launch, player data showed that some players were not engaging fully with the opening. Working with the director group, we reworked the onboarding and cut the tutorial roughly in half, so players reach the action sooner. The difficult part was reaching agreement to remove work the team had built with care.

2.3Directing

From 2022 I was Design Director. I led the design team, which grew to roughly twenty designers, through launch on seven platforms in 26 languages and managed eight direct reports. I met weekly with Mojang and Microsoft to review design progress and work through larger questions about the game's direction. After launch I worked with a small team on the updates, from new units and structures to nine monthly challenges.

3.1What I owned

Feature owner for core progression: DedSec (the hacker collective you play as in Legion) Upgrades, and operative skills and traits. I built the tuning spreadsheets and implemented the systems in engine. Later I led the overhaul of core progression, reworking costs and unlock rates across several connected systems. I also designed much of the onboarding in the tutorial level.

3.2Where progression lives

Legion's premise is that you can recruit and play as anyone in London, and with permadeath on, any of them can be lost for good. That removes the assumption most progression systems rest on, a single character who grows over the game. Growth can sit on DedSec, through upgrades every operative shares, or on the operatives themselves, through levels they earn by being played. My design used both, so the central tuning question was how much power each one held.

3.3The tuning problem

Power on the operative makes a veteran worth protecting. With too much of it, a single loss leaves missions out of reach, and the design assumed that is where players give up. Power in team upgrades is a floor every recruit gets, but with all of it there, a veteran is no stronger than someone recruited an hour ago. New recruits started at a level that rose with the DedSec level, so a player could recover from a loss. XP from recruiting rewarded growing the roster, though too much of it left behind players who kept to one favourite.

The model below follows that design. In the released game operatives do not level up, so growth comes from team upgrades and from who you recruit. The What shipped switch shows that version.

4.1What I owned

Feature owner for the core online systems: matchmaking, group management, proximity voice chat and emotes, from the rules through to the screens players use. Owning the system and its interface at the same time keeps the design honest, because the player only ever sees the waiting. I was also the UI and online designer on the Base of Operations feature team, and contributed design to the HUD, co-op features and the Dark Zone.

4.2Wait time vs quality

Matchmaking trades how long a player waits against how good the group is, where a good group is players close to your level. A search starts at the player's own level and widens the longer they wait. If nobody is found, the player starts the mission as host, and matchmaking keeps adding players to the group while they play. Every rule that keeps groups closer in level also shrinks the pool a player can match into. They have to hold up hardest when few players are online.

Every game settles somewhere on that line. It works better when the team chooses the point deliberately and can say why, rather than letting it fall out of the tuning. The model below is a single queue, with Solve showing my pick.