Voltron Data is an early-stage company creating high-performance data access and in-memory computing tools based on Apache Arrow to accelerate enterprise data analytics. We are a collection of open-source maintainers who have been driving open-source ecosystems over the last 15 years, particularly in the C++, Python, and R programming ecosystems.
We are assembling a global, diverse team to build a new foundation for data analytics with Apache Arrow. This foundation will usher in a wave of innovation in data processing that can take full advantage of the speed and efficiency offered by modern hardware.
We are looking for a highly motivated Senior or Principal Compute Kernels Engineer to join Voltron Data’s team. On the team, you’ll have the opportunity to help support and grow the Voltron Data and Apache Arrow ecosystems. You will work closely with Voltron Data development teams to help design, architect, and implement high-performance data analytics computational primitives.
What you will be working on:
Below is a rough timeline of where you can expect to be at different points during your career path starting in this position. 

Upon joining:

      • Spending time learning about the Apache Arrow memory layout, compute primitives, and APIs
      • Learning and embracing the Apache development process
In addition, Principal Engineers will be expected to:
      • Familiarizing yourself with the different partners for compute on Apache Arrow

Within a month:

      • Implementing new high performance vectorized compute kernels
      • Optimizing performance of existing compute kernels
      • Participating in peer code review in Compute Kernel development
      • Contributing to technical discussions and technical design documents
In addition, Principal Engineers will be expected to:
    • Reviewing SIMD Libraries for targeting multiple CPU architectures
    • Architecting / Designing performant compute kernels that use SIMD instructions
    • Leading technical discussions, building technical design documents, and reviewing work in Kernel development
    • Collaborating with other teams at Voltron Data to ensure that compute kernel APIs can be utilized as effectively and efficiently as possible
    • Collaborating with the open source Apache Arrow ecosystem to make Arrow the most performant data analytics solution possible

Within 6 months:

      • Developing a comprehensive set of low level benchmarks for to enable observing and addressing performance regressions
      • Ensuring that the compute kernels are compatible and performant across platforms (Linux, MacOS, and Windows)
      • Identifying and building reusable components and primitives across different functions to ensure a high quality and maintainable codebase
In addition, Principal Engineers will be expected to:
    • Architecting / Designing a plan to support multiple platforms with compute kernels (Linux, MacOS, and Windows)
    • Helping to balance development prioritization between development of new kernels and optimization of existing kernels
    • Designing process to ensure kernels are consistently improving in performance
    • Working with benchmarking team to ensure all performance concerns are addressed

Within 12 months:

      • Analyzing compute kernels implementations to identify inefficiencies and designing solutions to address those inefficiencies
      • Continuing to build out the collection of compute kernels to ensure they are as high quality as possible, balancing performance, usability, and maintainability
In addition, Principal Engineers will be expected to:
    • Acting as a compute kernels tech leader within Voltron Data
    • Defining standard methodologies for compute kernel development and promoting these across the Voltron Data and Apache Arrow ecosystems
    • Ensuring that the compute kernels are as high quality as possible, balancing performance, usability, and maintainability

Previous experience that could be helpful:

      • Building libraries in C++, especially using Modern C++
      • Building and/or using performance portability libraries on accelerator hardware
      • Building distributed algorithms and libraries
      • Building libraries that target network acceleration hardware
      • Building throughput optimized, IO-intensive libraries and/or applications
      • Writing and/or optimizing code using CPU SIMD intrinsics, CUDA, ROCm, SYCL, OpenCL, or other hardware performant languages 
In addition, Principal Engineers:
    • Using and/or building performance portability libraries on accelerator hardware

Additonal Information:

For NYC-based applicants, the expected salary range is $165k to $220K + equity + benefits.

*Note: Disclosure as required by NYC Pay Transparency Law

Actual starting pay will be based on job-related factors, including exact work location, experience, training, and skill level, so may be higher or lower than what is shown on this posting.


• Work from Anywhere - Payroll and Benefits in 150+ Countries
• Unlimited PTO
• Medical, Dental, and Vision
• Retirement [USA Only]
• Home Office Budget
• Continuing Education Budget
We are an equal opportunity employer and value diversity at our company. We do not discriminate on the basis of race, religion, color, national origin, sex, gender, gender expression, sexual orientation, age, marital status, veteran status, or disability status. We will ensure that individuals with disabilities are provided reasonable accommodation to participate in the job application or interview process, to perform essential job functions, and to receive other benefits and privileges of employment. Please contact us to request accommodation.

Apply for this Job

* Required
resume chosen  
(File types: pdf, doc, docx, txt, rtf)
cover_letter chosen  
(File types: pdf, doc, docx, txt, rtf)