Senior ML Engineer — Player Experience

Rafael Santos

Full-time · Senior · Los_Angeles

This role is no longer accepting applications

Similar open roles you can apply to now:

About the role

We are a mid-size games studio with 310 staff and three live service titles. Our largest title has 4.8 million monthly active players across PC and console. Live service games live and die on retention. Retention depends on whether the game feels personalised, fair, and progressively rewarding to each individual player — which means it depends, increasingly, on machine learning. Our player experience team builds the ML systems behind matchmaking quality, dynamic difficulty adjustment, personalised content recommendations, and churn prediction. These are not research problems. They run in production, they affect millions of play sessions per day, and the feedback when they go wrong is immediate — players post about it within minutes. We are looking for a Senior ML Engineer who has built and maintained real-time ML systems at meaningful scale, and who finds the specific challenges of player behaviour modelling interesting. The domain has unusual properties: data volumes are enormous, time-series signals are noisy and high-frequency, and the ethical dimensions of influence on player behaviour are not ones we treat lightly.

Responsibilities

Design and train ML models for matchmaking quality prediction, personalised content ranking, and churn risk scoring
Own the real-time inference pipeline from game client event to model prediction within SLA
Define experiment design standards for A/B testing model changes — control groups, metric selection, and statistical power
Work with game designers to ensure model behaviour aligns with intended player experience, not just engagement metrics
Mentor two mid-level ML engineers and contribute to technical direction for the player experience platform

Requirements

6+ years of ML engineering with production experience on real-time personalisation, recommendation, or player behaviour systems
Python at a senior level — you write the code that sets the standard for the rest of the team
PyTorch for training and evaluating sequential behaviour models and deep recommendation architectures
Kafka for real-time event stream ingestion from game clients at high throughput
Apache Spark for batch feature computation over player history and session data
SQL for exploratory analysis, feature validation, and ad-hoc investigation of player behaviour anomalies
Experience designing A/B experiments for ML model changes in a production game or consumer product environment

Benefits

$138,000 – $168,000 base salary + annual performance bonus
Full remote — our engineering team is distributed across North America and Europe
25 days PTO plus company holidays
$2,000 annual learning and conference budget
Game developer credits and lifetime access to our studio\'s titles — they are genuinely good games