At Akttyva Therapeutics, we are developing pioneering deep learning approaches for structure-based small molecule drug discovery. We leverage those technologies in one of the largest applications of machine learning in the life sciences to find treatments for vascular leak-underlined conditions.
We value a collaborative and transparent culture that fosters scientific and technical discussion. We strongly believe that evidence wins over opinions, and aim for evidence-based approaches throughout our programs from target selection to nomination of clinical candidate. Our team members have expertise in a wide range of disciplines-from computational chemistry and structural biology to clinical development, and we encourage cross-functional learning.
About the role
We are looking for a cheminformatics scientist with strong experience in compound modeling and data for target docking, ADMET and binding prediction to join our team. The team is relatively small, so there are lots of opportunities for both career growth and substantial contribution towards our success.
- Be a key contributor to our machine learning algorithms for compound discovery and lead optimization.
- Communicate well and work with other scientists and software engineers to build our late-stage optimization program.
- Help maintain Akttyva’s scientific presence with conference presentations and publications.
- Use your background and insights to collaborate across the R+D organization on a diverse range of projects.
- MS or Ph.D. in Chemistry, Physics or life sciences, with 0-2 years of industry and/or post-doctoral experience.
- Strong experience in navigating the interplay between data quality, quantity, diversity, statistics and machine learning model development.
- Basic understanding of ADME/Tox principles and experimental assays and their interpretation.
- Strong coding skills in at least one high-level programming language (Python, R, Java, C++, etc)
- Ability to work with people from diverse disciplines within cross-functional teams
- Familiarity with latest advances in ligand-based modeling and applications: graph-convolutional neural networks, GANs, multi-task models, transformers, variational auto-encoders, few shot learning, active learning
- Experience with real-world evidence for drug development projects