The Company
Dyno Therapeutics is reshaping the gene therapy landscape through AI-powered vectors. Through the application of our transformative technologies and strategic partnerships with leaders in gene therapy, we believe a future with life changing gene therapies for millions of people is within reach. Our team includes world-class molecular and synthetic biologists, protein engineers and gene therapy scientists working alongside software engineers, data scientists, and machine learning experts to transform the landscape of available gene therapy capsids.
Dyno has been named the 2021 NEVY Emerging Company of the Year, an Endpoints 11 Company and one of America’s Best Startups by Forbes in 2022 and 2023!
The Role
Machine Learning Intern. Dyno is the pioneer of using machine learning for black-box protein design. The Machine Learning team creates models and algorithms that power an ML-in-the-loop experimental platform with the goal of designing superior AAV capsids to unlock gene therapy delivery. Some of the tools the team has built to date include sequence-to-function predictive models, generative models for de novo sequence design, black-box optimization algorithms, and distribution shift detection methods.
Summer internships are full-time (40 hours per week) and run June - August with h
ybrid/remote working flexibility.
How You Will Contribute
As a Machine Learning Intern, you will be responsible for contributing to Dyno’s machine learning platform by helping to conduct research at the cutting edge of machine learning and protein engineering. Examples of the rich technical challenges that the ML team is tackling include coping with designed distribution shift, measurement noise, and experimental batch effects, and modeling useful unsupervised capsid representations. These examples are by no means exhaustive, and we welcome your ideas for testing new methods on Dyno’s abundant data!Responsibilities:
- Develop sequence-function or structure-function prediction models using state-of-the-art modeling architectures
- Design experiments to improve and validate our computational methods
- Reason about uncertainty and outliers in biological datasets
- Develop unsupervised methods for characterizing promising optimization targets
Who you are
- Team-oriented contributor excited to work as a trusted collaborator with other researchers
- Attentive to detail and able to independently work through challenges
- Keen on translating research insights into practical solutions
- Eager to learn about biology, and motivated to advance the state of ML knowledge in biology
- Strong execution skill and ability to operate in a fast paced environment
Basic qualifications
- Currently enrolled in a PhD program in Machine Learning, Protein Engineering, Applied Math, Statistics, Optimization, Computational Biology, or similar topics
- Track-record of researchin machine learning applications
- Expertise in Python scientific computing (numpy, scipy, pandas, Jupyter notebooks) and ML stack (sklearn and one of Pytorch or Tensorflow)
- Eager to learn about the intersection of machine learning and biology
Preferred qualifications
- Track-record of research in machine learning applications for protein science, protein design protein engineering, or similar topics
- Publications and/or presentations in leading Machine Learning or Biology conferences or similar
- Familiarity with structural protein design or training large language models
- Familiarity with any of the following is not required, but is helpful: Next-Generation Sequencing (NGS) data, molecular biology, structural biology, protein engineering, gene therapy, immunology, virology, and experimental techniques in high throughput assays
We are an equal opportunity employer and value diversity at our company. We do not discriminate on the basis of race, religion, color, national origin, gender, sexual orientation, age, marital status, veteran status, or disability status.Job Type: Full-time
