RESPONSIBILITIES: 

• Design, develop, and maintain a real-world deep learning framework that improves or expands upon Vilya’s macrocycle drug design pipeline.
• Collaborate and communicate with fellow experts in deep learning as well as the broader interdisciplinary Vilya team, including our research scientists in chemistry and biology.
• Get hands-on experience working with heterogeneous data gathered by our chemistry and biology teams, including a variety of in vitro assay measurements.

BASIC QUALIFICATIONS:

• Current PhD student in computer science, computational biology, bioinformatics, or a field applicable to computational drug discovery.
• Strong background in the fundamentals of machine learning, including linear algebra, calculus and statistics.
• Experience developing, deploying, and managing deep learning models and formatting large datasets for real-world problems.
• Proficiency in Python and PyTorch.
• Ability to work well with an interdisciplinary team and to communicate complex scientific ideas to diverse audiences.
• Availability to work for 12-16 weeks during Summer 2026
• Ability to work on site in our Seattle, WA office. No relocation costs will be provided through this internship.

PREFERRED QUALIFICATIONS:

• Experience with Slurm, Singularity, BigQuery and/or other cloud and high-performance computing infrastructure.
• Experience with concepts and deep learning techniques relevant to drug discovery and protein design, such as transformers, graph neural networks, equivariance, diffusion generative modeling, etc.
• Experience profiling and optimizing deep learning training/inference performance, including techniques such as pruning/quantization, parameter-efficient fine-tuning (e.g., adapters/LoRA), and implementing or integrating custom CUDA/Triton kernels or fused ops.
• Specific experience working with structure prediction or backbone diffusion models, such as AlphaFold, RosettaFold, Chroma, Boltz, etc.
• Experience working with reinforcement learning or active learning techniques, e.g. for optimizing molecules based on feedback from assays or expensive computational oracles.
• Experience with molecular property prediction, e.g. solubility, partition coefficients, toxicity, etc.
• Exposure to concepts in drug discovery such as ADME / Tox / DMPK.

VILYA BENEFITS:

• Opportunity to work in a disruptive startup with a talented, experienced, and dedicated team
• Monthly commuter stipend
• Salary: $10,000/month for the duration of the internship