# Machine Learning Scientist, RNA Design

Live page: https://affinitytalent.bio/jobs/machine-learning-scientist-rna-design

## Role summary

- Series A RNA therapeutics company
- Seattle, WA, US
- Hybrid · Senior
- Compensation: $190k - $222k + equity + annual bonus

Develop ML models that help design and prioritize RNA therapeutic constructs before wet-lab testing.

## About the role

The company is building programmable RNA medicines and wants to use experimental data more directly in construct design. You will build models that connect RNA sequence, structure, chemistry/design features, delivery/formulation context, and assay readouts to better construct-selection decisions. The model outputs need to be useful to RNA biology partners and strong enough to guide the next experimental cycle.

## Responsibilities

- Model development: Develop and evaluate sequence, structure-aware, or multimodal models that predict RNA construct performance, stability, expression, delivery behavior, or assay risk.
- Evaluation and validation: Design train/test splits, baselines, error analyses, prospective validation plans, and failure reviews that make model performance useful for real discovery decisions.
- Cross-functional translation: Work with RNA biology, screening, and data-engineering colleagues to turn assay data into training sets and model outputs into testable construct hypotheses.

## Requirements

- PhD with 3+ years, or MS with 5+ years, applying machine learning to biological, chemical, genomic, RNA, protein, or other molecular datasets.
- Strong Python experience with PyTorch, TensorFlow, JAX, or similar ML frameworks, including clean implementation and reproducible evaluation.
- Experience with sequence models, representation learning, generative models, structure-aware models, active learning, or multimodal biological data.
- Good judgment around train/test leakage, dataset shift, noisy labels, assay bias, and model evaluation for prospective experimental use.
- Ability to explain model behavior and limitations to wet-lab scientists and turn feedback into the next analysis or experiment.

## Technical stack

Python, PyTorch/TensorFlow/JAX, sequence modeling, RNA structure features, representation learning, ML experiment tracking, Docker, Git, RNA assay data.