Hedral Inc.

Role Description We are seeking a Machine Learning Intern to join our innovative team. This internship will focus on applying advanced computer vision, language models, and reinforcement learning to complex, domain-specific architectural and engineering data. The role offers the unique opportunity to work on cutting-edge problems at the intersection of AI, spatial reasoning, and the AEC industry, helping shape the future of automated design and structural analysis. Location: Austin, TX or New York, NY - Primary preference for Austin or NYC; remote-eligible for exceptional candidates with a proven track record. Core Responsibilities - Research and prototype computer vision models and language models tailored for domain-specific tasks, such as understanding and processing architectural plans, engineering documents, and building system data. - Develop robust data pipelines for curating, training, and fine-tuning models on diverse engineering data, including 2D drawings, 3D geometries, and text-based specifications. - Implement machine learning algorithms for tasks such as object detection, semantic segmentation, and advanced reasoning within the AEC domain. - Explore and implement reinforcement learning frameworks to optimize and automate complex decision-making processes in the built environment. - Collaborate with the engineering team to deploy AI models into our core design and analysis workflows, applying MLOps best practices for scalable machine learning deployment. - Conduct rigorous experiments and evaluate model performance on real-world AEC use cases to ensure scalability and accuracy. Qualifications - Currently enrolled in a graduate or undergraduate program in Computer Science, Engineering, Machine Learning, Applied Mathematics, or a related field. - Strong proficiency in Python, with a solid foundation in deep learning and hands-on experience using frameworks like PyTorch or TensorFlow. - Familiarity with core machine learning concepts and techniques, including supervised/unsupervised learning, model training and evaluation, and common architectures (e.g., CNNs, GNNs, transformers). - Demonstrated research experience, such as publications, preprints, conference submissions, or substantive research projects, with the ability to read, critique, and build on recent ML literature. Bonus Qualifications - Experience with any of vision, language, and graph neural networks or 3D/geometric deep learning, including CNNs (U-Net, ResNet), GNNs, and NeRFs. - Familiarity with modern vision, language generative models (e.g., VAEs, Diffusion, Transformers, ViTs, Multimodal models). - Knowledge of reinforcement learning principles and frameworks as applied to optimization or decision-making problems. - Background in Engineering, Architecture, or AEC with hands-on experience processing complex engineering data or spatial representations (CAD/BIM), and familiarity with relevant software (e.g., SAP2000, ETABS, Revit), reinforced concrete/steel design, and building codes (e.g., ASCE 7, ACI 318, AISC 360). - Experience with MLOps practices, including experiment tracking, model deployment, or working with large-scale datasets and distributed training. How we operate - Ownership & Intensity: We operate with high autonomy and a shared sense of urgency to redefine an entire industry. - Bias for Action and Efficiency: We prioritize automation over manual effort to solve high-consequence problems, shipping quickly and often. - Data-Driven Decisions: Hedral’s core advantage is not just automation, but that our system learns from real, stamped engineering work. Every project feeds back into the model. Benefits - Competitive Compensation and Benefits. - Environment for growth. - Engineering Redefined: Hands-on experience with the future of automated design that goes beyond traditional manual workflows. - Mentorship & Velocity: Fast-track career development working alongside industry leaders in both structural engineering and software development.