PhD Intern - AI Researcher for CAD (B-Rep) Generation

About the Role

Key responsibilities include finetuning LLMs such as Qwen 3 and gpt‑oss to generate CAD shapes. The intern will develop RAG and external‑memory methods to manage contexts exceeding 30,000 tokens that may contain previously generated or catalogued CAD models. Model outputs will be evaluated with standard CAD metrics, including B‑Rep validity and prompt adherence. Research findings will be documented, presented to internal stakeholders, and published when novel discoveries arise. • Build RAG and external‑memory pipelines handling >30k token contexts for CAD data. • Evaluate generated CAD outputs using B‑Rep validity and prompt‑following metrics. • Document research, present findings internally, and publish novel results. • Participate in a 16‑week paid internship (May 4 – Aug 21) with mentorship, tech talks, and flexible work options. • Contribute to projects that advance sustainable and innovative design solutions.

Key Responsibilities

• ▸rag pipelines
• ▸memory management
• ▸llm finetuning
• ▸cad evaluation
• ▸research publication
• ▸sustainable design

What You Bring

Applicants must be currently pursuing a Master’s or PhD in Computer Science, Artificial Intelligence, Machine Learning, or a related discipline. They need strong theoretical and hands‑on experience with deep learning and LLMs, especially finetuning frontier models, tool‑calling, and RAG. Proficiency in Python and collaborative development using Git is required. Familiarity with 3D topological data structures (e.g., half‑edge meshes, B‑Reps), geometry processing, and CAD frameworks such as OpenCascade is a strong advantage. Autodesk’s software powers sustainable buildings, clean vehicles, smart factories, and major entertainment productions, turning innovators’ ideas into reality. The company’s culture of belonging and diversity guides how employees collaborate, engage with customers, and impact the world. As an Autodesker, you can do meaningful work that helps build a better world for everyone. • Finetune large language models (e.g., Qwen 3, gpt‑oss) to generate CAD shapes. • Pursue a Master’s or PhD in CS/AI/ML with strong deep‑learning and LLM experience. • Demonstrate Python proficiency and collaborative Git workflow. • Possess knowledge of 3D topological structures and CAD frameworks (e.g., OpenCascade).

Requirements

• ▸llm finetuning
• ▸master’s/phd
• ▸python
• ▸git
• ▸opencascade
• ▸3d topology

Benefits

The 2026 Canada Internship runs for 16 weeks (May 4 – August 21) and is fully paid. Interns will contribute to meaningful projects, receive mentorship from industry leaders, and participate in tech talks and development activities. Autodesk’s Flexible Workplace approach supports office, remote, and hybrid work preferences. The program is designed to foster both personal and professional growth. Salary offers are based on experience, education level, and geographic location as part of Autodesk’s competitive compensation package. Autodesk is committed to diversity and belonging, cultivating an inclusive environment where everyone can thrive.

Work Environment

Hybrid