Member of Technical Staff, Kernel Engineer

We need to output HTML fragment with sections: About the role, Responsibilities, Qualifications, Compensation & Benefits (maybe Benefits). Also maybe "Nice to Have" could be part of Qualifications? The instruction: Use

for section headings (e.g. About the role, Responsibilities, Requirements, Qualifications, Skills, Benefits, Pay, Schedule). Include a section ONLY when posting has real, specific content. So we need sections: About the role (the paragraph after "About The Role"), Responsibilities (list), Qualifications (list), Compensation & Benefits (list). Possibly also "Nice to Have" could be part of Qualifications but we could make a separate "Nice to Have" section? Not listed as standard headings but could be considered "Skills"? But we can incorporate into Qualifications as a subsection? Better to keep Qualifications and maybe a separate "Nice to Have" as a subsection under Qualifications? The instruction says prefer a few substantive sections over many thin ones. So we could combine Nice to Have into Qualifications as a bullet list under "Nice to Have" maybe as a subheading? But we only allowed

for section headings. Could we have

Nice to Have

inside Qualifications? That would be another heading, increasing sections. Might be okay but they said prefer few sections. Could embed Nice to Have as a bullet list under Qualifications with a note? But we need to preserve all substantive information. We could include a

Nice to Have

as separate section; it's still substantive. However they said "Prefer a few substantive sections over many thin ones." Having Nice to Have as separate may be okay but adds another section. Could combine into Qualifications with a subheading using

? Not allowed? Only

for section headings. Could we use

Qualifications

and then inside have a

Nice to Have

? That would be using other heading levels, but instruction only says use

for section headings (e.g. About the role, Responsibilities...). It doesn't forbid other headings, but says use

for section headings. Might be safer to keep only

sections. Could we embed Nice to Have as a bullet list under Qualifications with a preceding

Nice to Have: ...

? That would preserve info without extra heading. Let's do that. Thus sections:

About the role

...

Responsibilities

• ... each bullet from responsibilities list.

Qualifications

• ... required qualifications bullets.

Nice to Have: ... list.

maybe as a paragraph with bullet list? Could do

inside? But we need to keep list formatting. Could do

Nice to Have:

... but that adds extra heading? Not needed. Better: after qualifications list, add a

Nice to Have:

then
list.

Compensation & Benefits

• ... each bullet.
We need to preserve all substantive information: The compensation paragraph includes base salary range, equity/stock options, medical/dental/vision, unlimited PTO, opportunities for significant technical ownership and impact, H-1B sponsorship. We'll list each as bullet. Also note "Actual compensation will depend on experience, technical depth, level, and role scope." Should be included. Now we need to strip boilerplate: The initial "About NeuroSpark" paragraph is not a section heading; we might include it as part of About the role? Actually there is an "About NeuroSpark" paragraph before "About The Role". The instruction says do NOT repeat the job title as a heading (the page already shows it) and do NOT put Job ID / requisition numbers as headings — fold minor metadata into a short

or drop it. The "About NeuroSpark" is company description; could be included as part of About the role? Probably we should include it as a paragraph under About the role? The "About The Role" heading already exists; we could put the company description as a paragraph before responsibilities? The instruction: "Use

for section headings (e.g. About the role, Responsibilities, Requirements, Qualifications, Skills, Benefits, Pay, Schedule)." So we should have an About the role section. The content for that section should be the description of the role, not the company overview. However we could include company overview as part of About the role? Might be okay but risk of deviating. The posting includes "About NeuroSpark" then "About The Role". The role description starts after "About The Role". So we should put the role description in About the role section. The company overview could be omitted? It's substantive but not about the role. The instruction says preserve ALL substantive information; do NOT invent or summarize away detail. So we need to keep the company overview somewhere. Where? Could be a preceding

before the headings, not as a section heading. That's allowed: we can have a short

for metadata. So we can put the company overview as a

before the first

. That satisfies preserving info. Thus output:

NeuroSpark builds and operates a high-performance AI inference platform that helps enterprises run large language models faster, cheaper, and at scale. Inference infrastructure is the foundation the entire AI application layer runs on — every AI product ultimately depends on how fast, how reliably, and how affordably models can serve their users. Our vision is to make that layer so efficient that compute is never the reason a good AI product fails.

About the role

NeuroSpark Inc operates an enterprise AI inference platform providing high-throughput, low-latency access to large language models across a distributed, heterogeneous compute fleet. As a Member of Technical Staff, Kernel Engineer, you will work at the lowest performance-critical layers of NeuroSpark's inference stack, designing and optimizing GPU and accelerator kernels that directly determine model latency, throughput, memory efficiency, and hardware utilization. You will work across the boundary between model architecture, GPU execution, inference runtimes, and distributed serving systems. The role involves identifying performance bottlenecks in real production workloads, developing hardware-aware optimizations, and integrating those improvements into the systems that serve models at scale. This is a hands-on individual-contributor role with significant technical ownership. You will work closely with engineers across inference, distributed systems, and infrastructure to push model performance across current and emerging accelerator platforms.

Responsibilities

• Design, implement, and optimize high-performance GPU kernels for performance-critical AI operations, including GEMM, attention, normalization, quantization, KV-cache operations, and Mixture-of-Experts (MoE) workloads.
• Develop and optimize kernels using technologies such as CUDA, Triton, C++, PTX, CUTLASS, and related GPU programming frameworks.
• Profile production inference workloads to identify bottlenecks across compute, memory bandwidth, memory hierarchy, kernel launch overhead, synchronization, and data movement.
• Optimize GPU execution through techniques including memory coalescing, shared-memory utilization, tiling, warp-level programming, Tensor Core utilization, operator fusion, latency hiding, and compute/communication overlap.
• Improve end-to-end LLM inference performance across latency, throughput, memory utilization, concurrency, and hardware efficiency, rather than optimizing kernels in isolation.
• Develop and optimize kernels for modern model architectures, including Transformer-based LLMs, attention variants, MoE models, and emerging model architectures.
• Implement and evaluate lower-precision execution and quantization strategies, including FP16, BF16, FP8, FP4, INT8, and other hardware-supported formats.
• Integrate optimized kernels and operators into inference frameworks and internal runtimes built around technologies such as PyTorch, Triton, vLLM, SGLang, TensorRT-LLM, or equivalent systems.
• Use profiling and performance-analysis tools such as Nsight Systems, Nsight Compute, PyTorch Profiler, roofline analysis, and internal benchmarking infrastructure to diagnose and resolve performance regressions.
• Build reliable benchmarks, correctness tests, and performance regression tests to ensure kernel improvements remain numerically correct and production-ready.
• Optimize workloads across multi-GPU and distributed environments, working with the broader infrastructure team on communication, parallelism, and compute efficiency.
• Help extend NeuroSpark's inference stack across heterogeneous hardware, including NVIDIA GPUs, AMD GPUs, and other current and emerging AI accelerators.
• Work closely with hardware vendors, inference engineers, and distributed-systems engineers to evaluate new accelerator architectures and translate hardware capabilities into production performance improvements.
• Contribute to architectural decisions affecting the Company's inference runtime, model execution layer, and hardware-performance roadmap.

Qualifications

• Strong experience in GPU programming, kernel development, high-performance computing, ML systems, or performance engineering.
• Proficiency in C++ and hands-on experience with CUDA, Triton, or another accelerator programming model.
• Strong understanding of modern GPU architecture, including: GPU memory hierarchy; threads, warps, blocks, and grids; shared memory and register usage; memory bandwidth and access patterns; Tensor Cores; synchronization and parallel execution; occupancy and instruction-level parallelism.
• Experience profiling and optimizing GPU workloads for latency, throughput, memory usage, and hardware utilization.
• Experience with performance-critical machine-learning operations such as attention, GEMM, quantization, KV cache, MoE, or other Transformer operators.
• Familiarity with PyTorch and modern ML inference or training execution stacks.
• Strong understanding of numerical correctness, floating-point behavior, and mixed-precision computation.
• Ability to reason from first principles about performance bottlenecks across hardware and software layers.
• Strong debugging skills and the ability to take performance work from profiling and hypothesis through implementation, benchmarking, and production deployment.
• Strong written and verbal communication skills and the ability to work effectively in a highly collaborative engineering environment.

Nice to Have:

• Experience with PTX/SASS, CUTLASS, CuTe, CUB, Thrust, or other low-level GPU libraries and programming abstractions.
• Experience with inference frameworks such as vLLM, SGLang, TensorRT-LLM, FlashInfer, or similar systems.
• Experience implementing or optimizing FlashAttention or other fused attention kernels.
• Experience with ROCm / HIP and AMD GPU architectures.
• Experience optimizing workloads across multi-GPU or multi-node systems, including NCCL and collective communication.
• Knowledge of ML compiler and runtime systems such as torch.compile, XLA, MLIR, TVM, or related compiler stacks.
• Experience with distributed training and inference, tensor parallelism, pipeline parallelism, or expert parallelism.
• Experience optimizing workloads on multiple accelerator architectures or developing hardware-portable kernels.
• Contributions to open-source projects in GPU kernels, ML systems, inference engines, compilers, or high-performance computing.
• Experience bringing new model architectures or accelerator platforms into production.

Compensation & Benefits

• Expected base salary range: $170,000 – $350,000 USD per year.
• Actual compensation will depend on experience, technical depth, level, and role scope.
• Equity / stock options.
• Medical, dental, and