Achira
Machine Learning Research Engineer (MLRE) - Workflows/Systems
About this role
Achira seeks a Machine Learning Research Engineer to design and maintain scalable distributed workflows for molecular ML experiments. You'll bridge research scientists and infrastructure teams while architecting systems that handle massive-scale data generation, training, and evaluation for foundation models in drug discovery.
What you'll do
- Design and maintain multi-stage asynchronous workflows for data generation, training, and model evaluation
- Architect ML systems with clean abstractions and scalable infrastructure
- Identify and resolve technical blockers at foundation model scale
- Serve as liaison between research scientists and infrastructure engineering teams
- Rationalize system design decisions and software architecture choices
- Develop observable, well-documented code with clear GitHub artifacts
What they're looking for
- PyTorch and JAX
- Asynchronous programming
- Distributed computing systems
- ML systems architecture
- Library and API design
- Workflow orchestration frameworks (Dagster, Flyte, etc.)
- Geometric deep learning or equivariant architectures (optional)
- Graph neural networks (optional)
Benefits
- Work at frontier scale with massive compute and data
- End-to-end ownership from ideation to deployment
- Hybrid work in San Francisco or New York City
- Collaboration with world-class ML researchers and scientists
- Well-funded, talent-dense organization
- Conference travel and corporate on-site opportunities
Opens the application — the Jobs AI extension fills it for you. Set up autofill
Opens the official application on the employer’s site. No login required.
Achira
Achira builds AI-driven drug discovery platforms using deep learning and foundation models for molecular simulation. The company is hiring Machine Learning Research Engineers and Software Engineers to optimize GPU-accelerated model implementations, design scalable distributed infrastructure, and manage large-scale ML pipelines across cloud environments.
View all jobs at AchiraLikely interview questions
- Walk us through a multi-stage asynchronous ML workflow you've built. What were the key design decisions, and how did you handle failure modes and retries?
- Describe your experience with PyTorch and JAX. In what scenarios have you chosen one over the other, and how do you think about framework trade-offs?