Machine Learning Engineer

Company DescriptionDrawScale develops AI software for construction takeoff and estimating workflows. The company is building systems that work directly from construction drawings, specifications, schedules, and related project documents to generate structured outputs for preconstruction use. DrawScale’s technology is focused on document understanding, information extraction, cross-document reasoning, and workflow automation for real-world construction environments.Role DescriptionDrawScale is seeking a Founding ML Engineer to help advance the core machine learning systems behind the product. This is a founding-level role for an engineer who is interested in applying machine learning to complex, document-heavy workflows and helping shape the long-term technical architecture of the company.The company has already developed substantial internal work around construction document understanding, information extraction, and workflow automation. This role is focused on improving and extending those systems into more robust production-grade capabilities that perform reliably on messy, real-world project documents.Construction documents present a difficult technical environment. Drawings often contain inconsistent scales, handwritten notes, incomplete structure, and domain-specific symbology. Specifications are dense, highly variable, and frequently reference standards or requirements that differ across project types and jurisdictions. This role is centered on building systems that can operate effectively in that setting and contribute directly to product capabilities used in practice.The Founding ML Engineer will work across the full machine learning lifecycle, including data pipelines, model development, evaluation, deployment, and production iteration. This is an engineering role focused on building reliable systems, not a pure research position.The role is remote initially, with a preference for candidates based in or open to relocating to the San Francisco Bay Area over time as the company builds toward in-person collaboration.What You’ll Do• Design and build document understanding systems, including layout analysis, entity extraction, and cross-document reference resolution• Develop and improve machine learning models for working with construction drawings, specifications, schedules, and related documents• Create evaluation frameworks, datasets, and data pipelines to support model development and measurement• Deploy models into production environments and monitor real-world performance• Work closely with product direction and domain input to understand workflows, edge cases, and failure modes• Contribute to the technical foundation, engineering standards, and long-term ML architecture of the companyQualifications• Strong machine learning engineering experience, including production system development• Experience in document understanding, computer vision, natural language processing, information extraction, or related areas• Strong Python skills and proficiency with modern ML tooling such as PyTorch, transformer-based frameworks, and supporting infrastructure• Demonstrated ability to take systems from prototype through deployment and iteration• Sound engineering judgment and the ability to make practical technical tradeoffs• Ability to operate independently in a fast-moving environment• Strong written and verbal communication skillsNice to Have• Experience working with construction documents, CAD/BIM data, or technical drawings• Background in document AI, multimodal systems, or information extraction• Experience building data pipelines, evaluation systems, and ML infrastructure at scale• Previous experience in a startup, founding, or early-stage engineering environment