Snabbfakta
-
- London
Ansök senast: 2024-12-31
Machine Learning Research Engineer
Publicerad 2024-11-01
About the Role
Join our leading-edge machine learning team as a Research Engineer, and play a pivotal role in pioneering machine learning solutions for genetic data interpretation. Our team is at the forefront of developing experimentally validated machine learning methodologies, transforming biobank-scale genetics data (from genotyping arrays to whole genome sequence) into actionable disease risk genes for drug discovery. Leveraging data from external biobanks like the UK Biobank (UKB) and our proprietary Osteomics clinical trial, we're unlocking new frontiers in genetics.
Responsibilities
- Collaborate with an interdisciplinary team to address complex challenges in data engineering, ML engineering, and software engineering.
- Architect and enhance data processing and loading systems to support training and inference for large-language models up to billions of parameters
- Develop and maintain research and production code bases, enabling rapid experimentation and efficient analysis of experimental outcomes.
- Lead initiatives to scale our capabilities for distributed training and improve software architecture for high-velocity research.
Requirements
- A Bachelor’s degree in Computer Science or a related quantitative field, with upwards of 4 years of relevant industry experience. Exceptional candidates with less experience but a strong technical foundation will also be considered.
- Proficiency in Python and Pytorch, with a solid understanding of data structures and computational complexity
- Experience on building out models from research to deployment
Demonstrable skills in at least two of the following areas:
- data engineering (e.g., handling billion-row data frames and databases),
- ml engineering (e.g., distributed training, optimization, compilation)
- Reinforcement learning
- Transformers experience
- Variational autoencoders
- software engineering (e.g., OOP, system design, CI/CD and testing)
- Comfort with industry-standard cloud infrastructures
- Knowledge of biology, genetics is an advantage but not essential
&v=024){kind=logo}
