Machine Learning Scientist

As a Machine Learning Scientist at Picarro, you will be an integral part of the AI Core team, focused on AI innovation, experimental modelling, and scientific validation. Your mission will be to develop highly efficient, adaptable, and efficacious large models for various general-purpose applications in chemical metrology. You will play a crucial role in creating AI models from scratch and guiding their development through all stages to achieve the research objectives. Your responsibilities will include experimenting with advanced architectures such as CNNs, RNNs, transformers, autoencoders, etc., that are suitable for the research objectives. You will also be tasked with developing training strategies like self-supervised learning and few-shot learning, optimizing loss functions and metrics for performance, managing hyperparameter tuning, and debugging training failures. Additionally, you will work on optimizing training pipelines and developing reproducible training/evaluation pipelines. To excel in this role, you should have expertise in PyTorch/TensorFlow or other frameworks, strong Python skills including NumPy, SciPy, scikit-learn, and GPU acceleration (CUDA, cuDNN). Experience with ML experiment tracking tools like W&B, MLflow, and RL frameworks such as Stable Baselines3 and Ray RLlib will be valuable. A passion for building AI agents with superior capabilities is essential. Ideally, you should hold a Bachelor's or Master's degree in Data Science, Computer Science, or a related technical field, along with at least 4 years of experience in machine learning and deep learning. Previous roles within AI research teams and a background in chemometrics, spectroscopy, or analytical chemistry would be advantageous. Knowledge of networks involving very large matrix operations is a plus. Join us at Picarro, the world leader in timely, trusted, and actionable data using enhanced optical spectroscopy. Our solutions are utilized in various applications worldwide, ranging from natural gas leak detection to greenhouse gas measurements. Our innovative software and hardware are designed and manufactured in Santa Clara, California, setting the benchmark for precision, ease of use, and reliability. If you require reasonable accommodation to complete the application process or need assistance due to a disability, please reach out to Picarro, Inc. at disabilityassistance@picarro.com for support.,