Staff Data Scientist

FYUL is the engine that powers on-demand commerce at global scale. Formed in 2024 through the merger of Printful, Printify, and Snow Commerce, it brings together tech, talent, and infrastructure to help people turn ideas into beautiful products. FYUL serves solo creators to entertainment giants, powering merch connected with millions via advanced tech, premium production, and global reach.

• 8+ years of experience building, deploying, and scaling production ML systems with staff-level technical leadership.
• Expertise in modern search and recommendation systems including Two-Tower models, ANN vector indexes, Transformer rerankers, and learning-to-rank algorithms.
• Proficient in classical computer vision (OpenCV) and deep vision architectures such as CNNs, Vision Transformers, diffusion models, and multimodal LLMs/CLIP.
• Experience building closed-loop ML systems integrating business outcomes with model training.
• Proficient in Python, PyTorch, LLM orchestration and agentic frameworks like LangChain, LangGraph.
• Familiarity with production serving frameworks (FastAPI, event-driven async), MLOps tools (CometML, Flyte), and cloud data warehouses (Snowflake).
• Experienced with AI-assisted workflows using Gemini, Claude Code, and Claude Cowork for ideation and rapid development.
• Product maker mindset with autonomy in end-to-end ML product ownership including architectural decisions and frontend API integration.

Nice to have:

• Experience in e-commerce, two-sided marketplaces, or creator platforms.

Leading & scaling recommendation and search systems:

• Lead architectural direction for candidate retrieval, multi-objective ranking, and learning-to-rank models.
• Advance personalization engines using sequence embeddings of merchant history, trends, and sales signals.
• Integrate downstream merchant sales and conversion data into upstream model training.

Driving AI design generation & vision pipelines:

• Oversee image generation pipelines combining LLMs/multimodal models with deep learning architectures such as diffusion models, UNets, and Vision Transformers.
• Enhance classical computer vision using OpenCV alongside deep learning models like ResNet, CNNs, and YOLO for delayering and asset validation.

Driving end-to-end product innovation (maker focus):

• Act as product engineer with full technical autonomy across the stack including model selection, serving pipelines, and frontend workflows.
• Prototype and ship ML features rapidly using AI-assisted development tools (Gemini, Claude Code, Claude Cowork).

Technical leadership, governance & mentorship:

• Define evaluation frameworks (NDCG, MRR, visual quality metrics) and manage continuous A/B testing.
• Guide make-vs-buy decisions, API vs. self-hosted model trade-offs, and inference optimization.
• Set applied ML standards, mentor senior ICs, run design reviews, and promote measurable business impact culture.

FYUL is an equal-opportunity workplace committed to diversity and inclusion and hires based solely on qualifications, merit, and work experience.