Sanofi is one of the top pharmaceutical companies with more than 100,000 people in 100 countries. We recently built a multidisciplinary team focusing on using
ML- and physics-based methods to accelerate design and engineering novel protein therapeutics.

https://sanofi.wd3.myworkdayjobs.com/SanofiCareers/job/Framingham-MA/Senior-Data-Scientist–Machine-Learning-for-Biologics_R2617876-1

https://sanofi.wd3.myworkdayjobs.com/SanofiCareers/job/Framingham-MA/Data-Scientist–Machine-Learning-for-Biologics_R2617878-1

 

 

Overview:

At Sanofi Large Molecule Research Platform, we have a strong ambition to utilize
machine learning (ML)
and artificial intelligence (AI) in all aspects in biologics discovery and development, with comprehensive programs spanning from deep repertoire mining to de novo design. Additionally, we are open and
actively collaborate with leading biotech and academia. The new opening senior data scientist role is to seek a machine learning expert with
strong expertise in modern data analysis methods, including NLP, graph, and 3D geometry-aware models.
The successful candidate will work in an interdisciplinary team to apply cutting-edge computation, ML/DL, sequence-based and structure-based methods
to resolve challenges in real-world drug discovery. This is an exciting opportunity to contribute to the process of design and engineering revolutionary biologics, including
multivalent, multi-targeting molecules by leveraging large amount of proprietary data coming from our industry-leading high-throughput automation platforms.
The successful candidate
will gain deeper insight into drug development process and transform ML/DL algorithms into accelerating biologics discovery and development process,
and make impacts to patients’ life.

Your responsibilities include:

• Evaluate and develop state-of-the-art computational methods to decode biophysical and geometrical features from antibody-antigen datasets and create predictive models
  for engineering.
• Develop and apply complex machine/deep learning solutions to our high-content and high-quality proprietary datasets, as well as public datasets.
  
• Perform data querying and feature extraction to improve current workflow for antibody/nanobody engineering, including affinity modification, cross-reactivity engineering,
  liability risk prediction and mitigation, multi-specific antibody engineering, and
  de novo antibody design.
• Maintain a keen awareness of recent developments in data science, bioinformatics, and state-of-the-art of ML/DL algorithms, aiming to accelerate development of new computational
  algorithms.
• Effectively collaborate with colleagues with diverse scientific background, identify problems and opportunities, combine computational and structural analysis to support
  large molecule projects.

Basic qualifications:

• Ph. D. in related field such as Biostatistics, Physics, Computation Biology, Biomedical Engineering, Computer Science, Applied Mathematics, Structural Biology with at least 2 years of experience or Master with
  at least 5 years of relevant experience.
• Significant depth of expertise in
  ML/DL, hands-on experience with modern machine learning models, including Transformers, Graph NN, Recurrent NN, MLP etc.
• Sufficient in programing using deep learning libraries such as PyTorch, TensorFlow, and Keras. Experience with cloud computing, parallel computing, and/or supercomputing
  is expected.
• Track record of applying machine learning/ deep learning approaches to solve molecule-related problems.
  Familiarity with protein structure or sequence featurization and learned embeddings.
• Experience with database mining, big data, and large-scale virtual screening using Bayesian Optimization or Gaussian Processes
  is desired.
• Familiarity with Data Visualization tools/libraries and dimensionality reduction algorithms.

Preferred qualifications:

• Understanding of protein structure and protein-protein interaction. Experience with structural analysis and optimization of biochemical and biophysical properties, like
  thermodynamics, with protein design tools e.g. Rosetta, BioLuminate, or MOE, etc.
• Understanding of biologics R&D process is a plus.