Machine Learning Engineer

Role & Responsibilities:

• Research, develop, and deploy machine learning systems onto the Giuseppe platform for food science & technology applications.

• Brainstorm with Experimental Chefs & Food Scientists to find data and ML-driven solutions for problems arising with new product development. We develop products ranging from plant-based milks to plant-based meats, so there are always engaging challenges to solve.

• Lead new machine learning initiatives end-to-end, from proposal to deployment. This requires having strong research, project management & communications skills and the potential to lead a small team.

• Analyze and debug issues with currently deployed models & suggest improvements to data, algorithms, or workflows.

• Identify new datasets that could help improve the performance of current algorithms.

• Read & implement recent ML papers that are relevant to our scope of work (e.g. implement a self-supervised algorithm proposed in a 2020 ICML paper).

• Dive into the domain of food science: be interested in learning about proteins, volatile molecules & the food industry in general.

• Contribute in general to department and organization projects.

Requirements:

• BS/MS in Computer Science or a related field with a focus on machine learning

• 3+ years of working experience in an Applied Machine Learning group

• Strong background in software engineering, unit & functional testing (Python, unittest, nose)

• Strong data science analytical skills with a traditional data science and Deep Learning stack (Python, PyTorch, Pandas, NumPy, SciPy, Scikit-learn, Jupyter, Matplotlib, Hugging Face, LangChain, Polars, etc.)

• Prior experience with implementing & training Deep Learning models; prior experience with training models on cloud infrastructure on GPU/TPU

• Familiarity with model packaging & deployment (Docker, Kubernetes, Google Cloud Platform)

• In-depth knowledge in one or several of the following areas preferred: Representational Learning, Generative Models, Transformers, bandit algorithms & convex optimizers

• Authorization to work in the U.S. is required