Machine Learning Engineer 4

We are looking for a passionate architect with deep machine learning expertise who can contribute to building, optimising and deploying model pipelines to be used by Adobe's innovative and industry-leading products in imaging, 3D, vector and audio/video domains. As a Firefly model architect, you will join an outstanding team of ML and service engineers building the future of digital experiences. You will help set long-term technical strategy and work across multiple Adobe teams to architect the pipelines of the future foundation models at Adobe. You will be empowered to make meaningful architecture decisions and prioritise model optimisation experiments. You will have the opportunity to develop novel ideas to advance the creative world through features in Adobe's products and reach millions of people worldwide!
Responsibilities:
Help architect and optimise large-scale foundation model pipelines in generative AI.
Design and develop the GenAI backend services for Firefly, creating GPU-optimised, efficient model pipelines that power the generative AI features on the Firefly website, PPro, Photoshop, Illustrator, Express, Stock and other applications/surfaces.
Collaborate with outstanding applied researchers and engineers to bring ideas to production.
Provide technical leadership and mentorship for junior team members.
Explore and research new and emerging ML and MLOps technologies to continuously improve Adobe's GenAI engineering effectiveness and efficiency
Review and provide feedback on features, technology, architecture, designs and test strategies.
Requirements:
Master's or PhD in Computer Science, AI/ML or related fields or B. Tech and strong experience in AI/ML.
10+ years of experience.
3+ years as a lead/architect.
Excellent communication and technical leadership skills.
Experience in leading a large number of contributors on time-sensitive and business-critical GenAI projects.
Experience in state-of-the-art generative AI technologies, such as GAN, diffusion, and transformer models.
Strong hands-on experience with large-scale GenAI model pipelines and/or shipping ML features.
Strong collaboration skills.
Experience with working in a matrix organisation, driving alignment, drawing conclusions and getting things done.
Preferred experience:
Experience in training or optimising models. (CUDA, Triton, TRT, AOT).
Experience converting models from various frameworks like PyTorch and TensorFlow to other target formats to ensure compatibility and optimised performance across different platforms.
Good publication record in computer science, AI/ML or related fields.

Experience
7-10 yrs