The Mercedes-Benz AI Experiences team is seeking a Machine Learning Engineering Intern to join our ML Product Development Team for the Summer of 2025. You will work with our Mercedes-Benz Research & Development North America (MBRDNA) team based in Silicon Valley. The team focuses on applying cutting-edge Machine Learning technology in products that will shape the future of Mercedes-Benz vehicles with the goal of enhancing our customer's experience.

In this position, you will work independently and collaboratively with our expert teams to translate state-of-the-art research into scalable, real-world solutions, towards a progressive future in the automotive industry. This position offers a unique opportunity to work at the intersection of ML research and practical engineering, addressing challenges in model optimization, inference speed, and integration within automotive systems.

Job Responsibilities:

• Investigate and develop advanced techniques for small LLMs, including transformer architectures and synthetic data generation for robust training.
• Explore methods for LLM fine-tuning and optimization, ensuring models are both high-performing and efficient.
• Collaborate with cross-functional teams to integrate LLM solutions into hardware platforms.
• Implement optimization techniques such as quantization, runtime adjustments, and inference speed improvements.
• Work with runtime deployment tools such as ONNX and TensorFlow Lite to optimize model performance on target hardware.
• Develop and evaluate retrieval-augmented generation (RAG) strategies to enhance model performance in dynamic, unstructured data scenarios.
• Document experimental findings, contribute to internal technical reports, and support potential publication efforts in top-tier conferences.
• Participate in team discussions, code reviews, and agile development cycles to continually refine and improve deployment strategies.

Minimum Qualifications:

• Currently enrolled in MS/PhD program in CS, EE, Math, or a related field, with a strong focus on machine learning, deep learning, and natural language processing
• Proficiency in Python coding, shell scripting, and working within Linux environments
• Demonstrated experience in developing and training deep learning models, especially with transformer architectures and language models
• Extensive experience with deep learning frameworks such a PyTorch and Tensorflow
• Experience with runtime deployment and optimization tools, e.g. ONNX, TensorFlow Lite
• Basic understanding of hardware deployment challenges, including containerization tools like Docker
• Experience with cloud-based tools and platforms such as Azure, Databricks, and Apache Spark
• Knowledge of model optimization techniques such as quantization, inference optimization, and runtime performance enhancements
• Basic knowledge of MLOps practices, including experiment tracking and model versioning using tools such as MLflow
• Understanding of ML workflow: preparing the data, implementing and training ML models, evaluating results, deploying inference on different platforms
• Experience with git or other version control systems

Preferred Qualifications:

• Experience with synthetic data generation for ML applications
• Prior exposure to LLM fine-tuning and evaluation methodologies
• Hands-on experience with retrieval-augmented generation (RAG) systems
• Familiarity with processing unstructured data in real-world environments
• A record of publications or contributions to reputable AI/ML, CV, or NLP conferences and journals
• Curious, self-motivated, and excited about solving open-ended challenges at Mercedes-Benz

Benefits/Perks:

*PTO

*Sick Time

Additional Information:

The current hourly rate for this position is as follows and may be modified in the future: $28 (Undergraduate Students)/$32 (Graduate Students)