Staff Machine Learning Engineer - Generative AI
About this role
Engineers who thrive on building complex multimodal models and shaping technical strategy from the ground up will find a strong fit at Xometry. The Generative AI team focuses on transforming a global manufacturing marketplace by applying advanced machine learning to real-world industrial challenges, such as parsing technical drawings and reasoning across text, images, and three-dimensional data.
In daily work, the individual in this role will guide model development, establish technical direction, and deploy large-scale architectures into production using AWS. The position exists to scale the company's automated processing capabilities and bridge the gap between heavy industrial workflows and modern AI research.
What you will do
- Set the technical direction for the Generative AI team while establishing best practices for high-impact innovation.
- Shape the AI roadmap by identifying valuable opportunities to apply generative models across the digital marketplace.
- Develop and deploy large language models and generative systems for multimodal document processing and structured data extraction.
- Explore innovative methods to combine text, images, and three-dimensional data to deliver smarter user solutions.
- Create robust data pipelines and training workflows capable of handling massive, complex datasets.
- Train, optimize, and deploy models into production at scale using cloud infrastructure.
- Collaborate closely with engineers, product leaders, and business stakeholders to integrate AI capabilities into real customer workflows.
- Mentor teammates on advanced machine learning methods, model architectures, and engineering best practices.
What you need
- Hold a bachelor's degree in Computer Science, Machine Learning, AI, or a related field, with advanced degrees serving as a plus.
- Bring five or more years of experience in machine learning or data science with deep expertise in generative models, LLMs, or computer vision.
- Demonstrate a strong track record working with large-scale language and vision models including Transformers, GPT, and VLMs.
- Possess hands-on experience managing multimodal data types such as text, images, and 3D files.
- Show proficiency in Python and core machine learning libraries like PyTorch, TensorFlow, pandas, and NumPy.
- Maintain solid grounding in probability, statistics, and optimization techniques for generative modeling.
- Demonstrate experience deploying models using cloud microservice architectures, with a preference for AWS.
- Apply strong software engineering skills encompassing object-oriented programming, testing, version control, CI/CD, and Infrastructure as Code like Terraform.
Nice to have
- Advanced degree (M.S. or PhD) in Computer Science, Machine Learning, AI, or related field
- Background in manufacturing, supply chain, or related industries
Benefits
- 401(k) match
- Medical, dental, and vision insurance
- Life and disability insurance
- Generous paid time off including vacation, sick leave, floating and fixed holidays, and maternity and bonding leave
- Employee Assistance Program (EAP) and other wellbeing resources
Skills and tech stack
Questions candidates ask
What is the compensation for this role?
The estimated base salary range for new hires is $140,000 to $230,000 annually plus a bonus. Total compensation depends on factors such as job-related skills, relevant experience, and location.
What technical stack will I use?
You will use Python alongside key machine learning libraries like PyTorch, TensorFlow, pandas, and NumPy. Models and pipelines are deployed at scale using AWS and cloud microservice architectures.
Are there citizenship or residency requirements?
Yes, candidates must be a U.S. Citizen or Green Card holder due to ITAR compliance requirements. Xometry also participates in E-Verify.
What are the core focus areas for the machine learning team?
The team focuses on generative AI efforts ranging from multimodal document understanding and extracting structured data from technical drawings to reasoning across text, images, and 3D data.