Note: The job is a remote job and is open to candidates in USA. Astrix is seeking a highly motivated machine learning professional to support a biotech research organization focused on artificial intelligence and machine learning for drug discovery and molecular design. The role involves developing, training, optimizing, and deploying advanced machine learning models, data pipelines, and MLOps workflows for scientific and biomedical applications.
Responsibilities
- Design, develop, optimize, evaluate, and deploy advanced deep learning models, including large language models, multimodal transformers, and generative AI models
- Build and optimize scalable data pipelines supporting machine learning and scientific applications
- Optimize model training and inference for performance, scalability, and accuracy using multi-GPU and cloud-based infrastructure
- Develop and maintain MLOps workflows covering model deployment, version control, monitoring, reproducibility, and ongoing model performance
- Develop machine learning approaches that connect diverse datasets, including genomics, transcriptomics, imaging, molecular, and clinical data
- Partner with scientists and engineers across disciplines to translate innovative machine learning methods into practical applications for drug discovery, disease research, and biomedical applications
- Independently troubleshoot complex modeling, software, and infrastructure challenges and drive solutions from development through deployment
Skills
- B.S., M.S., or Ph.D. in Computer Science, Machine Learning, Computational Biology, Data Science, Statistics, Mathematics, or a related quantitative discipline
- Must be authorized to work in the United States without current or future employer sponsorship
- 1–5 years of relevant professional experience, including postdoctoral research where applicable
- Strong foundation in data structures, algorithms, software engineering, and computational problem solving
- Expert-level Python programming skills
- Extensive experience with deep learning frameworks such as PyTorch, JAX, or TensorFlow
- Strong debugging and software development skills, with the ability to independently diagnose and resolve complex technical issues
- Experience with large-scale or distributed model training, such as DDP, Ray, FSDP, or DeepSpeed
- Experience with model deployment technologies such as Triton or ONNX
- Experience working with cloud and/or GPU computing infrastructure
- Hands-on experience with geometric deep learning, molecular cofolding models, neural force fields, or related scientific ML approaches is highly preferred
Company Overview