Machine Learning Engineer with Computational Bio

Our client is a biopharmaceutical company committed to the discovery, development, and commercialization of novel medicines to meet serious unmet medical needs in oncology, inflammation, and autoimmunity. Their mission is to improve the lives of patients by advancing a diverse portfolio of large and small molecules through rigorous research and development. Their AI Innovations Institute is dedicated to leveraging cutting-edge technology and artificial intelligence to enhance their research capabilities and drive innovation in drug discovery and development.

Requirements:

• **4 years of experience in ML development with proven experience in designing and delivering ML models to stakeholders.**
• **Experience with Python.**
• Experience in cloud computing, particularly with AWS.
• **Experience in a variety of ML domains such as supervised and unsupervised modelling, working with text or image based datasets.**
• Experience in building models end-to-end with the use of tools such as pandas, numpy, scikit- learn, and PyTorch.
• Experience in a collaborative environment where AI/ML work adheres to software development coding best practices.
• Strong problem-solving skills and the ability to work independently or collaboratively.

Computational Biology:

• Experience applying AI/ML methods to biology problems.
• General familiarity with biological problem spaces.
• Ideally hands-on experience working with antibody engineering, else similar experience in more general topics such as sequence alignment, structure prediction, virtual screening, etc.

Nice to Have:

• Experience with container tools such as Docker.
• Experience with MLOps tools such as AirFlow, MLFlow or Kubernet.

Responsibilities:

• Take ownership of ML development for different projects:
  • Lead the execution of ML / DS projects from understand project needs to completing project deliveries. Our team collaborates with various groups within the company to support their DS/ML needs. Different projects will have different timelines and different expectations. The ideal candidate would be able to design and deliver on these projects with scope that can span from data analytics to deep learning model development. The ideal candidate must also have a mindset of “getting 80% by using 20%” which means designing or prioritizing work that delivers the most impact with the least amount of effort.
  • Ensure that technical execution is aimed towards solving business metrics and not model metrics. As the field of AI / ML continues to evolve daily, it is very common for some ML projects to be just passed to a model with proper understanding of how each technology works. This leads to projects fixated on model performance metrics that may not align with stakeholder requirements. The ideal candidate would not be reliant on the “latest and fanciest” tools or models but would have a good understanding of how and when to use these tools. Being able to communicate the methods used and how it aligns with the stakeholder’s expectations is important.
  • Automate testing, enforce code quality, and apply development best practices. As the team works on multiple projects at a fast pace, we encounter pitfalls in methods or code we’ve developed but haven’t tested rigorously. An ideal candidate would help establish standards for best practices in coding that would make experiments easier to replicate, reuse, and reviewed by other team members.
  • Collaborate with the team and be proactive with improvements. As the team builds and creates solutions for various groups within the company, we often forget to pause and understand gaps in our team’s knowledge or areas where we could improve. An ideal candidate would be able to collaborate with the team on strategies to improve our infrastructure while navigating project deadlines. An ideal candidate would also help empower other team members to take on tasks that contribute to the development of the ML deployment pipeline.
  • Be open to learning and open to teaching. The ML Engineer will be someone the team relies on for model development. As technology evolves and our tools and models change, the team must adapt to these changes efficiently. We don’t expect the ML Engineer to be an expert in model deployment, nor do they need a background in the life sciences to be effective in this role. However, an ideal candidate would be open to learning from others, just as we can rely on their expertise.