Director, ML Services Engineering

The Opportunity 

Firefly Foundry is a new business venture at Adobe — an enterprise managed-service offering for custom multimedia generative AI. The offering includes deep-tuned custom image, video, and 3D models built on each customer’s IP, paired with creative workflows for content production and VFX, deployed across new and existing Adobe surfaces, and surrounded by a media intelligence layer. The business has gained significant traction in Media & Entertainment (M&E), marketing, and consumer retail, and is rapidly expanding into adjacent verticals. 

We are hiring a Director, ML Engineering to own the engineering function behind Firefly Foundry’s model services at enterprise scale. This is a multi-faceted executive role with end-to-end accountability for how Firefly Foundry’s models are productionized, served, and operated for our enterprise customers — and for the engineering organization that delivers that capability. 

What this role owns 

You will define and implement the technical strategy that turns Firefly Foundry from a managed service for early-design partners into a platform capable of serving hundreds of enterprise customers concurrently — each with their own IP, tenancy boundaries, and SLAs — powering everything from franchise extensions and new IP development to modern GenAI workflows at production scale. Specifically, you will: 

• Set the engineering operating model for productionizing custom generative models across image, video, and 3D — including the architectural patterns that simplify pipeline construction and let us absorb a heterogeneous mix of internal and external models without linear engineering cost. 
• Own the unit economics of Firefly Foundry inference. Cost-to-serve, GPU utilization, and gross margin on the managed service are your numbers. 
• Define the tenancy and data-isolation architecture that lets us honor enterprise IP contracts under audit. 
• Drive the self-serve roadmap that broadens Firefly Foundry’s reach and value beyond hands-on engagements. 
• Represent Adobe engineering in C-suite and senior technical conversations with studios, brands, and global enterprises — including VPs of Production, CTOs, and Chief Digital Officers. 

Who you will partner with 

• Applied Science — to ensure inference quality matches the training environment, and to make prioritization calls on emerging techniques for multimedia  
• Firefly Foundry Studio — to translate ambitious creative visions into reliable, high-performance ML systems that transform how content is conceived, produced, and delivered, and into concrete roadmaps with clear milestones and success metrics. 
• Post-sales field organization — engagement managers and creative technologists in customer engagements, where you serve as the engineering leadership representative and educate the field on APIs and services. 
• AI Platform, and adjacent Adobe orgs — to negotiate shared infrastructure, accelerator capacity, and serving primitives at platform scale. 
• Strategic partners — including GPU vendors and hyperscalers, where capacity planning, roadmap alignment, and partnership economics are part of your remit. 

What you will do 

Build and lead the engineering organization 

• Lead a multi-team engineering organization of ML engineers and engineering managers; recruit, hire, develop, and retain senior technical and leadership talent, and build a culture of engineering rigor and delivery discipline. 
• Hiring at scale is a material part of this role — Firefly Foundry is growing rapidly, and sustaining scaling momentum depends on it. You will integrate top technical and leadership talent into the organization at the pace the business demands. 
• Establish the engineering bar, the bench, and the talent strategy that let Firefly Foundry sustain 10x growth in capability breadth and traffic without linear headcount growth. 
• Define the operating rhythm — goal-setting, exec reviews, and engineering reviews — that keeps a fast-scaling org coordinated. 

Define and own the technical strategy 

• Own the multi-year architecture for training and inference at scale: pipeline construction, data pipelines, evaluation frameworks, model lifecycle management, and accelerator utilization (CUDA, NCCL, and the wider GPU stack). 
• Set the strategy for fast model deployment, parallel pipeline operation at scale, tenancy/data isolation, and self-serve capability buildout. 
• Make the build-vs-buy and prioritization calls on emerging GenAI techniques in partnership with Applied Science, based on material improvements in capability, cost, or speed. 

Own production reliability and economics 

• Hold the line on production SLAs for orchestrated and deployed model services. 
• Own analytics and observability across every model pipeline — quality, latency, cost, and utilization. 
• Drive cost-to-serve down on a multi-year curve while expanding capability. 

Drive customer and partner outcomes 

• Represent engineering in technical customer engagements with enterprise customers — translating creative and business requirements into ML roadmaps, milestones, and success metrics. 
• Co-design scalable, cost-efficient serving for real-time on-set use cases and high-volume social content generation, in partnership with infrastructure and platform teams. 
• Steward the GPU vendor and hyperscaler relationships that underwrite Firefly Foundry’s serving capacity. 

What you bring 

Leadership scope 

• 10+ years in applied machine learning and ML systems, including 5+ years leading engineering organizations — with prior experience leading managers of managers. 
• Demonstrated success shipping generative AI products in production at enterprise scale. 
• Proven ability to operate as a peer to VP-level partners across product, science, infrastructure, and field organizations, and to represent engineering credibly in front of senior customer and partner executives. 
• Track record of building engineering benches, defining career frameworks for emerging roles, and developing leaders who themselves become directors. 

Technical judgment 

• Deep understanding of the modern generative model landscape (diffusion, transformers, VAEs, latent video models, control/adapters, or similar) — enough to make architecture, investment, and prioritization calls with confidence, in close partnership with Applied Science. 
• Strong intuition for the economics and engineering reality of large-scale inference: accelerator stacks, model optimization and quantization, and the tradeoffs between quality, latency, and cost. 
• Experience designing and operating ML systems end-to-end — data, training, evaluation, deployment, monitoring, and continuous improvement — at production scale. 
• Familiarity with high-resolution media pipelines (4K+ video, high bit-depth) or adjacent bandwidth- and latency-sensitive domains is a meaningful plus. 

Communication and judgment 

• Executive presence and communication skills — able to brief executives, customers, and partners credibly, and to work directly with creative, production, and business collaborators to turn ambiguous problems into clear technical plans. 
• Sound judgment under ambiguity — comfortable making decisions with incomplete information and revising as new information arrives. 

Education 

• MS or PhD in Computer Science, Electrical Engineering, or a related field, or equivalent practical experience building and leading advanced ML systems.

​#FireflyGenAI