Senior Research Infrastructure Engineer
H
HUDAI infrastructure
Secondary Locations: Singapore, Remote (Asia), Remote (North America); open to remote candidates, 70-80% overlap with either San Francisco or Singapore time zonesFull-TimeSenior
Salary105,000 - 195,000 USD per year
Apply NowOpens the employer's application page
Job Details
- Required Skills
- DockerPythonRESTful APIsLinux
Requirements
- Demonstrate strong software engineering fundamentals and proficiency in Python.
- Have a record of building maintainable systems used by others.
- Have owned backend services, data pipelines, developer tools, or infrastructure from design through deployment and operation.
- Use sound judgment about APIs, data models, asynchronous workloads, and reliability tradeoffs.
- Be comfortable with Docker, Linux, and cloud infrastructure.
- Be able to debug failures across application code and the systems it runs on.
- Work closely with researchers and translate experimental workflows and open-ended research needs into useful software.
- Show independent technical judgment in prioritizing work and making pragmatic architecture decisions.
- Working knowledge of reinforcement learning, including environments, rollouts, reward signals, and evaluation workflows, is beneficial.
- Experience building research infrastructure, ML pipelines, or tools that improve research-team speed is beneficial.
- Familiarity with distributed job execution, containerized environments, or high-concurrency workloads is beneficial.
- Experience improving experiment reproducibility, data lineage, or visibility into complex pipeline failures is beneficial.
Responsibilities
- Build reliable infrastructure for agent behavior research, including environment execution, trajectory collection, and evaluation at scale.
- Develop reusable tools, APIs, and pipelines for launching experiments, managing datasets and configurations, and inspecting trajectories and evaluation results.
- Turn research prototypes and experimental workflows into tested, documented software that researchers can reuse and extend.
- Design experiments to investigate model behavior, agent failure modes, and data quality issues.
- Improve orchestration, observability, and recovery from failed runs to make experimentation faster, more reproducible, and more cost-efficient.
- Translate evolving research needs into infrastructure priorities and architecture decisions with research and engineering partners.
- Own systems from implementation through ongoing operation.
View Full Description & ApplyYou'll be redirected to the employer's site
