Senior Machine Learning Engineer (Ads ML Efficiency)

Join Reddit as a Senior Machine Learning Engineer, where you will own end-to-end performance optimization initiatives for high-impact production ML workloads. You will diagnose ML performance bottlenecks, design and implement performance improvements, and support faster and safer model launches. You will also partner with model engineers and platform teams, build reusable performance capabilities, and contribute to technical direction while mentoring engineers.

• Responsable de l'optimisation des performances des charges de travail ML en production, de l'identification des goulets d'étranglement à la mise en production.
• Profilage et diagnostic des goulets d'étranglement des performances ML, conception et mise en œuvre d'améliorations.
• Collaboration avec les ingénieurs de modèles et les équipes de plateforme pour identifier les opportunités d'optimisation et valider l'impact.

• Strong software engineering fundamentals: Experience implementing, debugging, testing, and maintaining performance-sensitive systems, with proficiency in Python and relevant systems programming technologies
• Strong profiling and benchmarking skills: Ability to identify bottlenecks systematically, establish reliable baselines, and validate measurable end-to-end improvements
• Hands-on ML performance engineering experience: Demonstrated ability to diagnose and implement meaningful performance improvements in production or production-relevant ML workloads
• ML systems understanding: Familiarity with production ML frameworks and the performance trade-offs across model execution, runtime, memory, hardware, and infrastructure
• Deep expertise in at least one ML performance area, such as inference execution, ML runtimes, compiler integration, GPU/kernel optimization, memory systems, or training execution
• Production engineering judgment: Ability to balance performance improvements with numerical correctness, model quality, reliability, maintainability, and safe rollout
• Good customer and platform instincts: can solve concrete bottlenecks while keeping maintainability, adoption, and future reuse in mind
• Strong communication: able to explain tradeoffs clearly to engineers and partner teams
• Experience with PyTorch, including performance optimization using technologies such as Dynamo, Inductor, AOTI, torch.export, or custom operators
• Experience building performance benchmarks, profiling tools, regression detection, launch-readiness capabilities, or cost observability toolings for ML workloads
• Experience with distributed-training performance, including GPU utilization, communication, data loading, memory efficiency, or training step-time optimization
• Experience optimizing large-scale inference or training workloads, particularly in recommendation systems, ranking, search, or retrieval
• Experience working across ML modeling, serving, and platform infrastructure teams
• Experience with hardware-aware optimization, including batching, memory management, operator fusion, mixed precision, quantization, or accelerator efficiency
• Experience with inference runtimes, GPU execution, or performance technologies such as TensorRT, ONNX Runtime, CUDA, Triton, LLVM/MLIR, or equivalent systems
• Specialized compiler internals or custom CUDA/Triton kernel development are valuable but not required. We value demonstrated performance engineering depth, sound technical judgment, and measurable workload impact over familiarity with any particular tool or framework