Senior Research Platform Engineer

Research Platform Engineer

Distributed Systems | Research Infrastructure | High-Performance Computing

The Opportunity

• We're looking for a Research Platform Engineer who enjoys solving difficult infrastructure problems and building systems that make complex research possible at scale.
• This role sits at the intersection of distributed computing, platform engineering, and computational research. You'll work closely with highly technical users to understand how they run experiments, process large datasets, and develop new ideas. Your job is to turn those needs into reliable, efficient infrastructure that makes their work faster and easier.
• We're particularly interested in engineers who have gone beyond deploying existing frameworks and have actually built, extended, or fundamentally improved the systems underneath them.
• The problems here aren't always solved by adding more compute. Sometimes the answer is a better scheduling strategy, a smarter approach to data movement, or a change deep within a distributed framework. You'll have the freedom to explore those trade-offs and the responsibility to turn good ideas into production systems.

What You'll Work On

• Build and evolve distributed compute platforms. Design systems that support demanding, data-intensive workloads across large clusters and heterogeneous computing environments.
• Make research faster. Work directly with researchers and technical teams to understand bottlenecks, improve workflows, and develop infrastructure that expands what they can accomplish.
• Solve problems deep in the stack. Improve scheduling, task execution, resource allocation, data locality, and fault recovery. Know when to extend an existing framework and when a different architecture makes more sense.
• Rethink how data moves. Develop approaches to caching, storage access, serialization, and data distribution that reduce unnecessary work and keep compute resources productive.
• Improve performance at scale. Investigate bottlenecks across distributed applications, from inefficient task graphs and network overhead to memory pressure and infrastructure constraints.
• Build resilient execution environments. Improve workload isolation, deployment efficiency, cluster orchestration, and recovery mechanisms so complex jobs run predictably.
• Create better tools for technical users. Build APIs, automation, observability, and workflow capabilities that simplify how researchers interact with large-scale infrastructure.
• Own the full engineering lifecycle. Take ideas from early investigation and architectural design through implementation, production deployment, and ongoing optimization.

What We're Looking For

We're looking for someone with strong systems instincts, real engineering depth, and the ability to connect infrastructure decisions to the people and workloads they support.

You should bring:

• Significant experience building or meaningfully extending distributed computing systems, execution frameworks, or large-scale infrastructure.
• Strong programming skills in Python, along with experience in a systems-oriented language such as C++, Rust, Go, or Java.
• A solid understanding of concurrency, parallel execution, distributed coordination, and the performance characteristics of modern computing environments.
• Experience designing systems that remain efficient and reliable as workload complexity, data volume, and resource demands increase.
• Familiarity with distributed scheduling, workload orchestration, caching, storage architecture, or cluster resource management.
• The ability to investigate difficult performance problems, reason through architectural trade-offs, and implement solutions that hold up in production.
• Experience working closely with demanding technical users and translating their needs into improvements that benefit an entire platform.

Most importantly, you've built things. You can explain why a system was designed a certain way, where it struggled, what you changed, and how those changes improved the experience for its users.

Experience That Would Stand Out

You don't need experience with every technology below. We're more interested in the depth of your work than the number of tools you've encountered.

• Distributed execution: Experience developing or extending systems such as Ray, Spark, Dask, Flink, or similar execution engines, particularly around scheduling, task graphs, parallelism, or resource management.
• Cluster infrastructure: Knowledge of Kubernetes, container runtimes, distributed schedulers, or workload orchestration systems, including their underlying architecture.
• Data-intensive systems: Experience with distributed caching, object storage, columnar data formats, efficient serialization, data locality, or high-throughput data processing.
• Cloud and infrastructure engineering: Experience building compute platforms across cloud or hybrid environments, with an understanding of elasticity, resource efficiency, and infrastructure automation.
• Performance engineering: Hands-on experience with profiling, memory management, concurrency, network optimization, fault tolerance, or reducing overhead in large-scale execution systems.
• Research-oriented computing: Exposure to scientific computing, quantitative analysis, machine learning infrastructure, simulation platforms, or other environments where computational efficiency directly affects experimentation.

We'd also be interested in engineers who have contributed to the internals of open-source infrastructure projects or developed specialized systems to solve problems that general-purpose tools couldn't address effectively.

What Makes This Role Different

• This isn't a position focused on maintaining cloud services or stitching together infrastructure tools.
• You'll have the opportunity to work on foundational engineering problems where improvements to the platform can change how quickly complex ideas are tested and developed.
• The work combines the technical depth of distributed systems engineering with the practical challenge of building a platform that people genuinely want to use.
• You'll collaborate with researchers who push infrastructure in unexpected ways, work through problems that don't have obvious answers, and help shape the architecture of systems that will continue to evolve.
• There's room for thoughtful experimentation, meaningful technical ownership, and engineering decisions that extend well beyond individual applications.
• The right person for this role is equally interested in how distributed systems work internally and what becomes possible when those systems work better.