Principal ML Engineer

Electronic Arts is seeking a Principal ML Engineer to own systems that prepare and deliver multimodal game data for large-scale model training. The successful candidate will architect and build shared training input pipelines, define schemas and data contracts, and optimize data throughput and accelerator utilization.

Responsibilities:

• Architect and build shared training input pipelines from certified game capture to accelerators
• Define schemas, data contracts, encodings, compression, and standards for dataset versioning and quality
• Build deterministic loaders and checkpoint-safe iteration for failure recovery
• Design reproducible dataset mixtures and balanced batches across modalities and resolutions
• Benchmark and enforce throughput, data-wait time, and accelerator utilization
• Diagnose bottlenecks and intermittent failures across storage, network I/O, CPU, memory, and accelerators
• Integrate compilers, runtimes, optimized kernels, and hardware-accelerated media processing
• Lead training-pipeline architecture reviews and mentor engineers

Qualifications:

• 20+ years of development experience, including owning production systems for distributed-training input
• Expertise building custom PyTorch training and ingestion systems or adapting TorchTitan, JAX/XLA with MaxText, Megatron Core/NeMo
• Expertise in FSDP2/HSDP, tensor/context/pipeline parallelism, mixed-precision training, and optimization on accelerator platforms
• Pre-training experience and understanding of how model architecture, data representation, batching, and orchestration affect training-system design and efficiency
• Expertise building large-scale multimodal streaming or hardware-accelerated ingestion with Ray Data, Mosaic Streaming, WebDataset, NVIDIA DALI, NVDEC
• Deep knowledge of sharding, sampling, packing, caching, prefetching, deterministic data mixing, checkpoint-safe iteration, and elastic restart