URBN Staff Engineer, GenAI

Urban Outfitters•Philadelphia, PA

20h

About The Position

URBN is hiring a Staff Engineer (GenAI) to join the development of AI-powered visual experiences, with a primary focus on building and operationalizing image and video generation systems. This role involves owning the engineering side of multi-model generative pipelines, turning research prototypes into reliable, scalable services that power AI-first innovations across URBN's digital ecosystem. The engineer will collaborate with a cross-functional team including data scientists, UX designers, product managers, creative partners, and domain experts to deliver significant business impact. The focus is on making generative systems robust, efficient, and production-ready. The ideal candidate should possess deep software engineering fundamentals, comfort with agentic AI tooling, and sufficient generative AI fluency to manage prompts, orchestrate multi-step visual workflows, and assess output quality. This position is for someone energized by the intersection of software engineering and generative AI, who wants to build the infrastructure and tooling to transform cutting-edge image and video models into real products, helping to shape the future of intelligent, AI-native creative experiences at URBN.

Requirements

1+ year of hands-on experience building or operationalizing image generation systems (diffusion models, multimodal pipelines) in a professional applied context.
Working familiarity with the rapidly evolving landscape of image and video generation models and orchestration patterns.
Proficient with AI-powered development tools (e.g., Cursor, VS Code Copilot, Claude Code, or Gemini).
Experience designing agentic workflows or AI orchestration systems using LangGraph, ADK, A2A, CrewAI, or similar.
Strong proficiency in Python, with practical experience using PyTorch and/or Hugging Face for model serving, fine-tuning support, and inference optimization.
Strong background in scalable RESTful APIs, microservices architecture, and high-availability distributed systems.
Proficiency with Docker, container orchestration, cloud-native services, and cloud-based AI infrastructure.
Experience with Terraform or similar IaC tooling.
Dedication to CI/CD pipelines, TDD, and automated code quality standards.
Excellent communicator who can bridge technical, data science, and creative teams.
Comfortable with ambiguity and minimal oversight.
Bachelor's or Master's degree in Computer Science, Engineering, Mathematics, or a related quantitative field, or equivalent practical experience.

Nice To Haves

Experience with video generation models, or with fashion/retail/e-commerce applications of generative AI such as virtual try-on, AI product photography, or on-model image generation.
Experience building tooling and orchestration for model fine-tuning workflows (LoRA, DreamBooth, ControlNet, IP-Adapter, textual inversion), enabling data scientists to iterate quickly on adaptation techniques.
Background in computer vision fundamentals (segmentation, detection, embeddings) or LLM evaluation frameworks for generative outputs.
Experience with prompt engineering at scale, creative design tools (Adobe Photoshop, Firefly, Figma), or data pipeline orchestration (Airflow, dbt, or similar).

Responsibilities

Design, build, and optimize image and video generation pipelines (generation, inpainting, upscaling, style transfer, conditioning, post-processing) into production-ready, observable services with attention to cost, latency, and throughput.
Design and develop agentic workflows (ADK, A2A, LangGraph, or similar), MCP servers (FastMCP or similar), and microservices (FastAPI, GraphQL, or similar) to orchestrate and serve generative AI capabilities at scale.
Develop prompt management systems and structured prompting strategies for consistent visual output, learning on the job how to wrangle prompts for fashion-specific and virtual try-on use cases.
Engineer consistency mechanisms and quality gates in partnership with data scientists who own evaluation methodology.
Integrate multimodal and vision-language models into production workflows for image understanding, automated tagging, captioning, and quality pre-screening.
Collaborate with Product Designers, Product Managers, Data Scientists, and other Engineers to translate brand and business needs into scalable generative AI solutions.
Evaluate new technologies, models, and vendors through proof-of-concept studies.
Implement and maintain ML architecture, including pipelines and applications that enable training and inference of generative models in production.
Champion best practices in full-stack algorithm engineering.