Principal Data Scientist
Core
Architecting the long-term technical vision for AI/ML ecosystems, specifically focusing on media understanding, generation, and multimodal data foundations for global entertainment metadata.
Role type
Principal Data Scientist (AI/ML Architect)
Builds
Multimodal machine learning systems, data foundations for petabyte-scale annotation/processing, and high-performance inference systems for real-time/batch workloads.
Domain
Media & Entertainment, Generative AI, Computer Vision
Deliverable
production ML models
Required skills
Python, Java, C++, PyTorch, TensorFlow, object detection, image segmentation, video understanding, distributed computing (Spark, Flink), MLOps (Kafka, Airflow, Kubernetes), cloud AI services (AWS Bedrock, SageMaker), GPU optimization (TensorRT, quantization, pruning)
Preferred skills
Experience with LLMs, diffusion models, vision transformers, multi-agent architectures
Technologies
PyTorch, TensorFlow, YOLO, Mask R-CNN, NVIDIA DeepStream, Triton Inference Server, TensorRT, Spark, Flink, Kafka, Airflow, Kubernetes, AWS Bedrock, SageMaker
Responsibilities
Shape technical vision for AI/ML across the content product ecosystem; Architect complex multimodal ML systems integrating visual, audio, and textual data; Lead design of horizontal foundational capability layers; Oversee development of high-performance inference systems; Define rigorous evaluation frameworks leveraging A/B testing and human-in-the-loop reviews; Mentor data science and ML engineering communities.
Seniority
Principal, strategy & mentorship
