Machine Learning Engineer

Machine Learning Engineer What you can expect As a Machine Learning Engineer on our R&D team in Germany, you’ll turn ideas into impactful product features. You’ll work on advanced AI challenges such as speech-to-text, translation, and large language model training that drive real innovation across Zoom’s products. You’ll design, train, and deploy models that advance the state of the art and shape the future of communication technology. Your work will directly influence how millions of people connect and collaborate every day. You’ll be part of a collaborative and inclusive team at Zoom’s Karlsruhe Center of Excellence for Multilinguality, where research meets real-world impact. About the Team The Speech Processing and Machine Translation team at Zoom is dedicated to making global communication seamless. This friendly team of experienced researchers develops and deploys advanced multilingual AI systems. They combine cutting-edge AI research, scalable engineering, and user-centered innovation to build state-of-the-art speech and language processing systems. Their goal is to deliver high-quality, multimodal, real-time, and offline translations for both speech and text across Zoom's products. The team collaborates closely with various departments and teams globally within Zoom to achieve their mission. Responsibilities Designing, training, evaluating, and optimizing machine learning models for speech and language processing tasks. Developing scalable ML pipelines for large-scale data processing and model deployment. Handling, curating, and synthesizing training and testing data while automating pipelines for efficient data processing. Implementing advanced evaluation frameworks for translation scenarios, including streaming speech-to-text and speech-to-speech translation. Working with advanced speech synthesis systems, customizing and integrating them into existing systems to enhance performance and user experience. Collaborating with researchers and engineers to integrate models into production environments. Experimenting with new techniques in deep learning, LLM fine-tuning, and model alignment to drive innovation and improve system capabilities. Analyzing model performance to identify areas for continuous improvement. Contributing to research initiatives through prototyping, experimentation, and publishing where applicable. Maintaining high standards of code quality, reproducibility, and technical documentation. Engaging in cross-functional projects with global teams to deliver innovative AI solutions to market. Supporting collaboration and knowledge sharing across research and engineering teams. Ensuring systems and processes align with organizational goals for scalability, efficiency, and innovation. Continuously refining methods to enhance model accuracy and deployment efficiency. What we’re looking for Bring 2+ years of experience; or a Master’s/Ph.D. in Computer Science, Artificial Intelligence, Machine Learning, or related fields. Demonstrate depth and breadth in state-of-the-art machine learning methods. This includes neural network training, LLM training, and post-training techniques (e.g., preference optimization and reinforcement learning with human feedback). Consider experience with speech transcription, machine translation, speech translation or speech synthesis as advantageous, though not required. Show advanced coding skills in Python, C/C++, or Java. Apply hands-on experience with machine learning model development using frameworks such as PyTorch, TensorFlow, or JAX. Work with large-scale text and/or audio data processing and distributed systems. Use proven mathematical knowledge and understanding of machine learning and statistics. Contribute effectively within a team environment. Bring the following, which would be nice to have: experience building deep learning models in PyTorch. Train and evaluate multilingual or multimodal LLMs, and optimize distributed training on GPUs. Also, communicate clearly in writin ... (truncated, view full listing at source)