← All work

Mobile Premier League · 2020–2024

Building a design system for a portfolio of games

Scaling game craft, platform UX, and a multidisciplinary design organisation without turning quality into a bottleneck.

GamesDesign leadershipPlatform
RoleHead of Design
Timeline2020–2024
Team25+ product designers, game designers, artists, tech artists, and animators

Measured impact

01

6M+ daily active players across India, the US, Brazil, and Nigeria

02

26 game titles shipped across the portfolio

03

Generative-AI content pipelines cut production time by about 90%

The challenge was not one game

MPL was a real-money gaming and interactive platform operating across very different markets. Every new title brought its own interaction patterns, art pipeline, production risks, and business model. Local optimisation was easy; portfolio coherence was the hard part.

I led design across the platform and games, building a 25+ person multidisciplinary organisation spanning product design, game design, game art, tech art, and animation. The goal was to create enough shared infrastructure that teams could move quickly, without sanding away what made each game distinct.

From craft functions to one operating system

We moved away from disciplines behaving like downstream service teams. Instead, we organised around player journeys and portfolio-level problems, with craft leadership supporting shared standards and growth.

Three decisions mattered most:

  1. Shared patterns, not cloned games. We standardised high-friction platform moments — onboarding, wallets, tournaments, game launch, recovery — while protecting the core fantasy of each title.
  2. Observability for design. We created ways to connect experience quality, technical performance, and funnel behaviour so critique could move beyond taste.
  3. Reusable production capability. In-house game art, tech art, animation, and design systems reduced dependencies and made iteration part of the normal release loop.

AI as a production system, not a demo

I led the move toward generative-AI workflows for content and asset pipelines. The useful work was less about a single model and more about designing the system around it: prompts, review gates, asset consistency, failure handling, and human judgement. In the right parts of the pipeline, this reduced production time by roughly 90%.

Scale did not come from adding more process. It came from making quality easier to repeat.

The outcome was a team able to support a large live platform, ship a broad portfolio of games, and still make specific, high-quality decisions close to the player.

Continue exploring

Different medium.
Same builder instinct.

More work ↗Enter the lab ↗