PhD Intern, Machine Learning: MSI, Product Operations Team

Job Description:

Imagine what you could do here. At Apple, new ideas have a way of becoming phenomenal products, services, and customer experiences very quickly. Every single day, people do amazing things at Apple. This position involves a wide variety of skills, innovation, and is a rare opportunity to be working on groundbreaking, new applications of machine learning, research, and implementation. Ultimately, your work would have a huge impact on billions of users across the globe. You can help inspire change, by using your skills to influence globally recognized products' manufacturing.\",\"description\":\"The goal of Apple's Manufacturing and Operations team is to take a vision of a product and turn it into a reality. Through the use of statistics, the scientific process, and machine learning, the team recommends and implements solutions to the most challenging problems. We're looking for ML researchers to help us revolutionize how we manufacture Apple's amazing products. Put your skill \u0026 passion to work in this highly visible role.\",\"responsibilities\":\"Research, build \u0026 adapt ML approaches for domain adaptation of multimodal foundation models to Apple manufacturing. Domain adaptation here is broad - spanning modalities including text, images, and video, such as generating physics-consistent videos adapted to Apple manufacturing\\nDevelop techniques that steer large foundation models without modifying their weights, and multimodal knowledge-graph reasoning - domain-specific Q\u0026A, multi-hop reasoning, disambiguation, and tool/retrieval orchestration over manufacturing knowledge\\nTrain, evaluate, and analyze model performance (e.g., exact-match/F1 accuracy, explanation quality, and generation fidelity) on public benchmarks and on internal datasets where applicable\\nApply and extend LLMs, VLMs, and World Models to advance domain adaptation for manufacturing\\nCollaborate within Apple to apply solutions to industrial problems\\nConduct presentations of analyses and results to a wide range of audiences including executives\",\"preferredQualifications\":\"Familiarity with applying Machine Learning, especially Computer Vision in a Manufacturing / Assembly setting \\nFamiliarity with multimodal understanding \u0026 reasoning, knowledge-graphs \u0026 generative video models\\nPrior publications in relevant field\",\"minimumQualifications\":\"Pursuing PhD in Computer Science / related field\\nDeep understanding of Machine Learning concepts, Probability and Statistics\\nUnderstanding of Computer Vision \u0026 VLMs, with demonstrated skill in training models and applying them to ML application development.\\nAbility to prototype research ideas as working code (Python)\\nAbility to meaningfully present results of analysis in a clear and impactful manner