Machine Learning Engineer, Autodesk

• BSc or MSc in Computer Science, or equivalent industry experience
• 3+ years of professional experience in training, deploying and optimizing large and/or generative models
• 8+ years proficiency with modern deep learning techniques (e.g. Network architectures, regularization techniques, learning techniques, loss-functions, optimization strategies, etc.) as well as frameworks (e.g. PyTorch, Lightning, Ray etc.)
• Experience and intuition for optimizing inference performance for large-scale deployments
• Experience debugging production distributed systems
• Experience in end-to-end deployment and maintenance of ML Models on cloud services and architecture (e.g. AWS, Azure)
• Excellent written documentation skills to document code, architectures, and experiments

Desirable

• The work we do at Autodesk touches nearly every person on the planet
• By creating software tools for making buildings, machines, and even the latest movies, we influence and empower some of the most creative people in the world
• As a Principal Machine Learning Engineer with a focus on model optimization, you will play a critical role ensuring high performing models that enable our customers imagine, design, and make a better world
• You are an experienced Machine Learning Engineer who is passionate about solving problems and building things
• You are excited to collaborate with AI researchers to implement generative AI features in Autodesk products
• You will report to a research manager in the Autodesk AI Lab team within Autodesk Research
• Optimize new and existing ML models deployed in Autodesk software
• Quantize, distill, and prune large-scale models and evaluate their efficiency and effectiveness
• Utilize methods such as model compilation, PEFT and others to deliver and serve models rapidly
• Experiment with various techniques to analyze as well as optimize throughput, latency and efficiency of productionized models
• Optimize models for maximum utilization of compute resources
• Evaluate and integrate new promising methods from recent AI/ML literature for faster inference
• Present results to collaborators and leadership