Principal ML Architect - End-to-End Systems & Data

Adobe Inc.; San Jose (CA); On-site; USD 206,000 - 379,000; Full time

Job summary

Adobe Applied Science & Machine Learning (ASML) is seeking a Principal Scientist, ML - Overall Architect to serve as the cross-cutting technical owner bridging our training and inference framework, ML model building, and data across Adobe's next-generation video and image foundation models.

The role requires authority over end-to-end architecture, leadership of large-scale distributed training and inference systems, and a strong drive for principled co-design of models, data and infrastructure.

Qualifications

• PhD in Computer Science, Electrical Engineering, AI/ML, or equivalent depth demonstrated through research contributions or principal-level systems impact.
• Demonstrated ability to reason authoritatively across distributed training systems, inference/deployment infrastructure, model architecture, and data pipelines.
• Hands-on expertise with large-scale distributed training (PyTorch FSDP, tensor and pipeline parallelism, checkpointing, fault tolerance) and inference/serving for large generative models.

Responsibilities

• End-to-End Architecture Ownership: Own the overall technical architecture spanning the Training & Inference Framework, ML model building, and data.
• Training & Inference Systems Leadership: Direct large-scale distributed training and inference infrastructure for performance, reliability, and cost efficiency.
• Model Architecture & Data Co-Design: Drive interaction between model architecture choices, training recipes, and data design.
• Cross-Team Technical Direction: Connect TIF, model building, and data teams to unblock seams and align interfaces.
• Architecture Decisions for Next-Generation Models: Own architecture-level decisions for GenRender6 / Gen6.5 and GenEdit1.
• Performance and Scalability: Translate scaling and deployment requirements into concrete architecture decisions.

Skills

Distributed training

Inference systems

Model architecture

Data pipelines

Cross-team leadership

Senior-level judgment

Education

PhD in CS/EE/AI-ML

Tools

PyTorch FSDP

Tensor parallelism

Pipeline parallelism

Checkpointing

Inference serving

Job description

Adobe Applied Science & Machine Learning (ASML) is seeking a Principal Scientist, ML - Overall Architect to serve as the cross-cutting technical owner bridging our training and inference framework, ML model building, and data across Adobe's next-generation video and image foundation models.

The role requires authority over end-to-end architecture, leadership of large-scale distributed training and inference systems, and a strong drive for principled co-design of models, data and infrastructure.