Computational Chemistry Co-op

Who we are:

Kymera is a clinical-stage biotechnology company pioneering the field of targeted protein degradation (TPD) to develop medicines that address critical health problems and have the potential to dramatically improve patients’ lives. Kymera is deploying TPD to address disease targets and pathways inaccessible with conventional therapeutics. Having advanced the first degrader into the clinic for immunological diseases, Kymera is focused on building an industry-leading pipeline of oral small molecule degraders to provide a new generation of convenient, highly effective therapies for patients with these conditions. Founded in 2016, Kymera has been recognized as one of Boston’s top workplaces for the past several years. 

How we work:

• PIONEER: We are courageous, resilient and rigorous in our mission to improve patients’ lives through our revolutionary degrader medicines.
• COLLABORATE: We value trust + transparency from everyone. Our goals are shared, our decisions data-driven and our camaraderie genuine.
• BELONG: We recognize our differences, inviting curiosity and inclusivity, so that our people are valued, seen, and heard.

How you’ll contribute:

In this role you will leverage advanced computational techniques to construct ML models capable of accurately predicting the potency of novel compounds, utilizing both 2D and 3D molecular descriptors. Key aspects of the project will include data preprocessing, molecular docking, descriptor calculation, model development, and rigorous validation, ensuring the robustness and reliability of the resulting predictions. The scope of the project may evolve in alignment with company goals and priorities.

• Initial Data Set Curation and Preprocessing
  • Create clean internal compound bioassay datasets (intern development goal: learn importance of data curation)
  • Internalize public benchmark datasets for the prediction of ligand binding affinity
  • Select compounds with a range of potency, removing ambiguous data and poor performing compounds (intern development goal: learn the types of assays for potency and the limitations of each assay)
• Ligand Docking
  • Create reasonable docking poses for each compound in the set (intern development goal: learn about tools used to predict protein and ligand poses)
  • Evaluate the calculated energetics of ligand poses and capture descriptors that represent the ligand-protein interactions of docked ligand poses
• Model Development and Preliminary Testing
  • Calculate the 2D and 3D descriptors of the ligands, including ligand-protein docked pose descriptors
  • Develop initial 2D and 3D ML models for the ligands (intern development goal: learn how to build ML models and split data to accurately quantitate predictiveness in prospective application and ultimately yield the best possible models)
• Model Validation
  • Evaluate model performance, and refine the model when possible (intern development goal: Master the process of model validation and learn about overfitting)
  • What descriptors are the key contributions to the model?
• Compare Model to other available tools
  • Utilize in-house and public benchmark datasets to assess model performance in comparison to other methods, e.g. MM-GBSA (intern development goal: understand how new tools can impact timelines and deliverables, and how to discern whether a model should be used in an active project)
  • Identify scenarios where the new modeling methodology performs better/worse than others

Skills and experience you’ll bring:

• Currently enrolled in M.Sc. or higher degree program in Biotechnology, Computer Science, Data Science, Chemistry, Pharmaceutical Sciences, and Drug Discovery, or related fields.
• Strong proficiency in Python programming, including experience with libraries such as Pandas or NumPy.
• Experience building and evaluating machine learning models for real-world applications.
• Knowledge of the life sciences, with a preference for a background in chemistry, is highly desirable.
• Excellent problem-solving skills and the ability to work independently as well as collaboratively within a team.
• Strong communication skills, both written and verbal, with an ability to explain technical concepts to non-experts.
• Familiarity with deep learning techniques in the context of chemistry (e.g., graph neural networks, molecular embeddings, etc.).
• Knowledge of cheminformatics concepts and tools (e.g., molecular descriptors, fingerprints, cheminformatics libraries such as RDKit).
• Knowledge of databases and chemical data repositories (e.g., PubChem, ChEMBL).

** This is a co-op position July 2025- December 2025 **

Skills and experience not an exact match?:

Go ahead and submit your resume (and a cover letter, if you’d like!).  If this role isn't right for you, we can keep you in mind for future opportunities.

Kymera Therapeutics is proud to be an equal opportunity employer, seeking to create a welcoming and diverse environment.  All applicants will receive consideration for employment without regard to race, color, religion, creed, gender, national origin, age, disability, veteran status, pregnancy, sex, marital status, gender expression or identity, genetic information, sexual orientation, citizenship, or any other legally protected class.