Destava

Member of Technical Staff — Inference-Multi-Hardware

radixark is hiring for Member of Technical Staff — Inference-Multi-Hardware in Palo Alto. Staff level. First seen November 7, 2026; open as of November 19, 2026. Jobs taken straight from company career pages — no aggregator in between.

Save to applyFree forever. No commitment, no card required.

Company
radixark
Where
Palo Alto
Type
Full-time
Level
Staff
First seen
November 7, 2026
Listed via
Ashby
Open for
11 days, as of November 19, 2026

About this job

RadixArk is looking for a Member of Technical Staff to enhance AI inference and training efficiency across a range of hardware, such as NVIDIA and AMD GPUs, Google TPUs, server CPUs, and new accelerator types. The position entails kernel migration, creating hardware abstraction layers, and engaging with silicon vendors on upcoming platforms. This broad technical role reports directly to the engineering team based in Palo Alto.

Responsibilities

  • Initialize inference and training systems on novel accelerator platforms and achieve competitive performance levels
  • Develop hardware abstraction mechanisms enabling one codebase to perform efficiently across vendors without branching
  • Migrate and refine kernels across diverse programming models and memory architectures
  • Construct cross-platform benchmarking, profiling, and regression tools to verify performance metrics on all targets
  • Resolve numerical inconsistencies and correctness discrepancies between different platforms
  • Collaborate with vendor engineering teams on unreleased hardware, compiler and driver challenges, and roadmap input
  • Coordinate with kernel, runtime, distributed systems, and product engineers to secure end-to-end performance gains
  • Act as the internal authority on each platform's specific strengths and capabilities
  • Return hardware-specific optimizations, benchmarks, and portability efforts to open-source SGLang and Miles

Requirements

  • Over 4 years in systems, performance, or ML infrastructure roles
  • Advanced proficiency in one accelerator programming model (CUDA, ROCm/HIP, Pallas/XLA, Triton, or vendor SDK), with proven capacity to quickly master additional ones
  • Solid grasp of accelerator architecture, including memory hierarchy, bandwidth constraints, occupancy, and their trade-offs
  • Proven background in crafting or refining high-performance kernels for ML applications
  • Hands-on experience with distributed execution and communication libraries (NCCL, RCCL, MPI, or similar)
  • Strong command of C++ and Python
  • Robust system-level debugging and profiling capabilities, including on platforms with incomplete or unreliable tooling
  • Demonstrated history of performance improvements successfully deployed in production

Nice to have

  • Experience in bringing ML workloads to new silicon platforms
  • In-depth practical knowledge across multiple vendor ecosystems
  • Familiarity with compiler stacks (XLA, MLIR, TVM, Triton) or developing compiler passes and IR transformations
  • Experience in designing hardware abstraction layers or portable kernel interfaces
  • Quantization and mixed-precision work across various numeric formats and hardware support levels
  • Experience with distributed inference systems (SGLang, vLLM) or training/RL frameworks (Miles, Megatron, veRL, TorchTitan)
  • CPU inference optimization using AVX-512/AMX, oneDNN, or NUMA-aware execution
  • Experience optimizing collective communication at scale or scaling workloads to over 1000 accelerators
  • Contributions to open-source kernel, compiler, or ML systems
  • Direct collaboration with silicon vendors or cloud partners on technical assessments
  • Background in HPC or other performance-critical systems

Benefits

  • Stock options

Read more and apply on radixark's site

Questions about this job

When was this Member of Technical Staff — Inference-Multi-Hardware job at radixark posted?

Destava first saw it on November 7, 2026, on radixark's own careers page (Ashby).

Is radixark still accepting applications for it?

It was open on radixark's own site when Destava last read it, on November 19, 2026, 11 days after we first saw it. When the company takes it down, this page says the job has closed.

How do I apply for Member of Technical Staff — Inference-Multi-Hardware at radixark?

On radixark's own site, at jobs.ashbyhq.com: open the posting. Destava does not take the application. Save the job to your Destava board to track it, see how well you fit and tailor your résumé to it.

Is this radixark job posting legit?

Destava read it on radixark's own careers page at jobs.ashbyhq.com, run on Ashby, not on a job board. First seen November 7, 2026. Apply on the company's own site, and never pay to apply for a job.

What other jobs is radixark hiring for?

radixark has 19 open postings on its own site: see all jobs at radixark.

How many computer hardware engineers jobs are open in San Francisco Bay Area?

807 open, 208 of them new this week, at 158 companies: see them all.

Applying to this one?

See how well you fit, before you apply

Save it to your board. Add your résumé once, and Destava shows where you are strong, what is missing and how to pitch yourself, then tailors your résumé to this posting.

  1. Found it
  2. Save it
  3. See your fit

Free forever. No commitment, no card required.

Or apply on radixark's own site

With Destava: Save to apply · Tailor my résumé · Write a cover letter

This page includes information from the O*NET 31.0 Database by the U.S. Department of Labor, Employment and Training Administration (USDOL/ETA). Used under the CC BY 4.0 license. O*NET® is a trademark of USDOL/ETA.