Machine Learning Scientist, Wayfair

• We are looking for a strong Machine Learning Scientist with deep technical expertise and a proactive, action-oriented mindset
• If you are passionate about working on cutting-edge AI problems, building scalable machine learning systems that solve real-world challenges, and directly contributing to significant business impact, this is the role for you
• PhD with 0-1+ years of experience or Master’s in Computer Science, Machine Learning, or a related quantitative field with 2+ years of full-time industry experience in applied research
• Proven track record of delivering successful machine learning projects from conception to production, demonstrating strong deployment, problem-solving, and maintenance skills
• Deep understanding of both traditional Machine learning approaches and deep learning techniques, reinforcement learning, and multi-modal understanding
• Deep understanding of LLMs, generative AI, and fine-tuning techniques, and Retrieval-Augmented Generation (RAG) techniques, with hands-on experience using models like GPT, Gemini, Claude, etc
• Deep understanding of data engineering concepts with experience in building scalable data pipelines for collecting, processing, and transforming data
• Professional coding expertise in languages like Python and Go, proficiency in SQL, and experience with data visualization tools; skilled in using ML frameworks (TensorFlow, PyTorch) and implementing CI/CD, containerization, and version control best practices
• Familiarity with ML Ops, cloud infrastructure, and engineering best practices (Airflow, Kubeflow, MLFlow, Kubernetes, Spark, Python, SQL)
• Excellent communication skills, with the ability to clearly articulate complex AI concepts to non-technical stakeholders while collaborating across teams
• Demonstrated ability to quickly learn new tools and techniques in a fast-paced, evolving environment, while managing multiple priorities with a high level of attention to detail and staying current with the latest ML research

Desirable

• The Catalog Science Foundations team at Wayfair drives growth by tackling fundamental challenges ranging from multi-modal product understanding, product relationships modeling, to GenAI-driven catalog intelligence at scale
• Our team provides the core capabilities that are crucial for enhancing the customer browse experience (e.g. , visual search, recommendations), streamlining supplier product onboarding (e.g. , product classification, image and content tagging), informing catalog competitiveness, and ultimately delivering significant impact for our customers, suppliers, and Wayfair
• In this role, you will be instrumental in developing and refining the advanced models and systems that power our core mission
• You will contribute to a diverse range of initiatives by applying cutting-edge deep learning techniques, vision-language models, and Generative AI agentic workflows
• You will work in close collaboration with a highly capable, cross-functional team to tackle complex, high-impact challenges and help pioneer innovative solutions
• Research and experiment with state-of-the-art multi-modal understanding techniques and algorithms. Design and implement evaluation strategies applied to real-world scenarios tailored to Wayfair use cases
• Leverage and fine-tune LLMs (e.g., OpenAI GPT, Google Gemini, Anthropic Claude, Open Source) to build AI-driven classifiers, product taggers, and quality control mechanisms
• Develop and refine the visual search system leveraging cutting edge technologies in computer vision, vision language models, and the large scale data orchestration
• Implement AI-powered automation for product data structuring, attribute extraction, and metadata validation—ensuring our catalog remains accurate, complete, and scalable
• Collaborate with top AI research and industry leaders (e.g., Google, Anthropic, Snorkel AI) to explore cutting-edge techniques in LLMs, data labeling automation, and scalable ML workflows
• Develop agentic AI workflows for automated schema definition, dataset generation, production relationship modeling, and LLM-based judgment systems to validate catalog data
• Partner with cross-functional teams across engineering, scientists, and product to ensure AI solutions integrate seamlessly into catalog systems
• Optimize cost, efficiency, and scalability of AI models, leveraging parameter-efficient fine-tuning (LoRA, QLoRA), knowledge distillation, and hybrid ML approaches