Artificial Intelligence Machine Learning Engineer

Key Skills

Aartificial intelligence, AIOps, Computer Vision (CV), CUDA kernels, Dask - Python library, Data science techniques, DataOps, Deep Learning, Google Cloud Platform (GCP), JAX framework, Keras software library, Linux Operating system, Machine learning techniques, MLOps tools, Natural Language Processing (NLP), Python Programming, PyTorch, R Programming, TensorFlow

Job Description

AIOS is part of the department serving some of the most prominent research and clinical programs at the Institute, from basic to translational research, to clinical deployment, and operationalization. The AIOS group encompasses expertise in AI, data science, machine learning, computer vision, NLP, production deployment, cloud infrastructure, data engineering, project management standards, and data labeling. Dana-Farber Cancer Institute (DFCI) provides expert, compassionate, and equitable care to children, adults, and their families, while advancing the understanding, diagnosis, treatment, cure, and prevention of cancer and related diseases. DFCI trains new generations of clinicians and scientists, disseminate innovative patient therapies and scientific discoveries around the world, and reduce the impact of cancer, while maintaining a focus on those communities who have been historically marginalized.

Located in Boston and the surrounding communities, Dana-Farber Cancer Institute is a leader in life changing breakthroughs in cancer research and patient care. We are united in our mission of conquering cancer, HIV/AIDS and related diseases. We strive to create an inclusive, diverse, and equitable environment where we provide compassionate and comprehensive care to patients of all backgrounds, and design programs to promote public health particularly among high-risk and underserved populations. We conduct groundbreaking research that advances treatment, we educate tomorrow's physician/researchers, and we work with amazing partners, including other Harvard Medical School-affiliated hospitals.

Responsibilities

PRIMARY DUTIES AND RESPONSIBILITIES:

• Meeting and consulting scientists requiring machine learning & AI support and designing plans and solutions.
• Delivering results for projects on-time and on-budget.
• Working as part of the broader team to identify longer-term solutions that will improve quality, speed and efficacy of our current projects and programs.
• Evaluating and benchmarking new libraries; prototype and pipeline development.
• Contributing to existing software packages; ability to identify needs based on client requests and convert those into tangible software designs
• Implementing best practices for software development (e.g. clear code/documentation, version control, unit testing) in all of their work.
• Communicating ideas and findings clearly to experts from various areas (e.g. pathology, radiology, data engineering).
• Keeping up to date with the state-of-the-art in computer vision and communicating findings from the literature to the team, wider department

Qualifications

Minimum Education:

Bachelor’s degree required.

Minimum Experience:

1 year of relevant experience. Deep machine learning & AI skills, at the interface with computer science. Python experience is required; R experience is a plus. Experience within a clinical or research environment preferred, experience with radiology and/or pathology imaging modalities is preferred (e.g. CT, MRI, H&E)

• KNOWLEDGE, SKILLS, AND ABILITIES REQUIRED:
• Excellent communication and effective problem-solving skills, track record in serving a variety of diverse customers and projects.
• Ability to work independently, prioritize, and manage people if needed, within an environment with ever changing priorities.
• Experience with one of the following:
• Natural language processing or Computer vision technologies, Transformers, Adversarial / Generative models, JAX + Flex / Haiku, Vision Transformers, Graph Networks, Federated learning, AutoML, Self-supervised learning, Causal ML, Reinforcement learning, Infrastructure as Code, DataOps (versioning, lineage, and governance), AIOps & MLOps life cycle (from deployment to monitoring to retirement), explainable AI, batch/online/streaming/edge training/inference, fully reproducible and auditable ML practices, CI/CD for large language models and large vision models, Multi-Cloud & Hybrid data platforms, productized Docker/Spark/Kubernetes solutions such as Databricks and Snowflake, High-throughput big data processing under redundancy / low-latency requirements, at least one deep learning framework (e.g. PyTorch, Tensorflow, Keras), distributed programming (e.g. CUDA, DASK), GCP, Linux, Git, Version Control, Commonly used medical image analysis software (e.g. 3DSlicer, CellProfiler, QUPath).