Job Description:

Architect multi-model systems combining skill, preference, trust, and safety signals for fair and meaningful matchmaking
Develop models for skill inference, player behavior prediction, trust & safety signals, and multi-objective optimization across fairness, latency, and experience quality.
Build and optimize real-time inference systems for personalized content, store offers, matchmaking, and player interactions at global scale.
Drive adoption of advanced modeling approaches including contextual bandits, reinforcement learning, graph ML, and session-aware personalization.
Partner with Data Engineering and Product to shape data schemas, feature pipelines, telemetry standards, and model observability across the ML lifecycle.
Define Responsible AI standards and implement fairness audits, bias mitigation, transparency, and safety mechanisms for matchmaking and social systems.
Lead post-launch evaluations of algorithmic impact on player sentiment, community health, and ecosystem stability.
Set organization-wide standards for model optimization (latency, throughput, memory), multi-model orchestration, and drift detection.

Required Qualifications:

10+ years in ML/Applied AI; 3+ years in principal/staff-level technical leadership.
Experience with large-scale, real-time ML systems (recommendations, personalization, matchmaking).
Expertise in graph ML, RL, and representation learning.
Proficiency in PyTorch, TensorFlow, JAX, and modern data/serving tools (Ray, Kafka, Flink, Redis).
Strong grounding in A/B testing, experiment design, and experience metrics.
Track record of setting ML strategy and standards across teams.

Desired Qualifications:

Professional background in gaming
Familiarity with Vertex AI, SageMaker, or internal large-scale inference systems.

Principal Machine Learning Engineer

Key skills