Principal Data Scientist - Gen AI

Xometry operates a digital marketplace connecting buyers with manufacturing capacity and resources. The company is seeking a Principal Data Scientist to lead Generative AI initiatives, including model development, multimodal document understanding, structured data extraction, and the integration of AI into the Siemens ecosystem. The role owns technical roadmaps, drives cross-functional machine learning projects, and mentors scientists and engineers.

Responsibilities:

• Own and drive an 18-24 month technical roadmap that balances innovation with disciplined business value delivery. Lead strategic planning and roadmap development for generative AI initiatives, identifying high-impact projects and aligning them with Xometry’s business objectives
• Develop and deploy generative AI models and large language models (LLMs) for multimodal document processing, focusing on extracting structured data from technical drawings, purchasing orders, and other complex documents. Lead the exploration and development of innovative text and image-based data processing solutions, including training and fine-tuning generative and language models
• Provide technical leadership to the Generative AI team, setting technical direction, defining best practices, and ensuring the team follows industry standards in AI and ML development. Champion the adoption of better approaches across teams and functions
• Influence the roadmaps, priorities, and resourcing of partner engineering, product, and business teams. Collaborate with cross-functional teams, including engineering and business teams, to align generative AI solutions with business needs and drive impactful applications
• Serve as the technical architect for the embedded DFM AI + IQE track, leading the integration of Xometry’s proprietary AI directly into the Siemens ecosystem, including Designcenter and Teamcenter. This work involves operating directly on native 3D geometry to enable real-time manufacturability feedback and pricing, creating a 'science fiction speed' digital thread from ideation to delivery
• Design and implement efficient workflows for data preparation, cleaning, and augmentation to support the training of generative AI models. Utilize cloud platforms (e.g., Amazon Web Services) for large-scale data processing, model training, and deployment. Continuously experiment and iterate on model performance, tuning architectures and parameters to improve accuracy and efficiency in a fast-paced, agile environment
• Serve as a respected subject matter expert and mentor to scientists and engineers across the organization. Mentor and guide team members on advanced machine learning techniques, model architecture design, and problem-solving strategies to elevate the team’s technical capabilities. Stay updated with the latest research in generative AI, deep learning, and multimodal data processing