Principal / Senior Principal Cheminformatics Machine Learning Scientist

We are looking for an experienced Senior Cheminformatics Machine Learning Scientist, who specialises in AI/ML and has a keen interest in small molecule drug design, to join our Cheminformatics team. The ideal candidate for this role will have demonstrable expertise in QSAR modelling and the development of AI/ML methods for chemistry.

The Cheminformatics team is a high performing cross-functional team that seeks to apply their knowledge to a diverse range of programmes from Target Identification through Hit ID, Hit Expansion and Lead Optimisation. Our role is to aid the advancement of our small molecule Drug Discovery programmes by devising computational solutions to project-specific challenges and applying new and existing technologies to support the needs of our wider portfolio. 

As a Senior Cheminformatics Machine Learning Scientist within the team, you will contribute to this challenge by leading our QSAR modelling initiatives within cheminformatics, to advance our small molecule drug discovery programmes. You will work closely with cheminformaticians, and medicinal and computational chemists, to develop QSAR models, design new AI/ML approaches for project-specific challenges, and apply a range of new and existing technologies to support the needs of our wider portfolio. 

• Lead our QSAR modelling initiatives and find new AI/ML-driven solutions to apply across our small molecule drug discovery projects.
• Build, evaluate and deliver QSAR models to advance our small molecule Drug Discovery programmes, and to support their use by project teams.
• Develop processes, customisable workflows and computational techniques that can be adapted and applied across the drug discovery portfolio.
• Contribute to the development of our technical cheminformatics capabilities, particularly in the area of applied AI/ML, and help define the long-term strategic thinking of the Chemoinformatics team.
• Collaborate and communicate effectively with members of the Chemoinformatics, Computational Chemistry, Bioinformatics, Drug Discovery, Artificial Intelligence, Engineering and Product teams.
• Nurture talent at BenevolentAI by sharing experience and offering a mentoring role, where appropriate.

Requirements

Essential Skills:

• PhD or equivalent in a field related to Chemoinformatics or Machine Learning, or a closely related field.
• Demonstrable experience in developing QSAR models for drug discovery, particularly in medicinal chemistry.
• Innovator of new ideas and approaches in the field of AI for chemistry, as demonstrated by appropriate papers, presentations, and code contributions to open source projects.
• Strong knowledge of Python, deep learning frameworks (e.g. TensorFlow, PyTorch), and state-of-the art ML approaches.
• Strong and demonstrable programming and technical skills, and familiar with open source and proprietary Chemoinformatics libraries e.g. RDKit or other leading industry toolkits.
• Practical experience processing and deriving novel insights from large chemical data resources, e.g. ChEMBL, SureChEMBL, and PubChem.
• A solid understanding of Chemoinformatics approaches and their application to live drug discovery projects, and being able to objectively design scientifically-merited experiments.
• Excellent communication skills, especially when working with colleagues from other specialities.

Desired Skills:

• Familiarity with the drug discovery process, and an understanding of what is involved in medicinal chemistry optimisations.
• Familiarity with computer-aided drug design approaches, such as compound library design, docking, virtual screening, molecular fragmentation, structure-based drug design, pharmacophore generation and analysis, multi-parameter optimisation.
• Familiarity with commercial Cheminformatics and computational chemistry tools, such as those from Schrodinger, ChemAxon, and KNIME.
• Familiarity with modern software development paradigms, including containerisation with Docker, GitOps, and cloud computing on AWS with Kubernetes.