Senior AI Inference Performance Engineer (Remote)

DigitalOcean; Denver (CO); On-site; USD 191,200 - 239,000; Full time

Job summary

DigitalOcean is seeking a Senior Engineer 2 to drive performance optimizations for AI inference across GPU architectures, focusing on memory bandwidth, throughput, and low-latency inference at scale.

You will lead benchmarking, optimize attention layers, and champion quantization techniques, while mentoring peers and aligning with product goals to deliver high-performance, developer-friendly AI infrastructure.

Qualifications

• 5+ years of experience in high-performance computing or AI infrastructure.
• Gen AI literacy with major model families.
• Hands-on experience with attention-layer optimizations and distributed GPU environments.
• Comprehensive understanding of NVIDIA and AMD GPU architectures and their software ecosystems (CUDA, ROCm, etc.).
• Open source experience: building with and contributing to OSS projects.
• Systems design skills in low-level GPU programming, optimization, memory access patterns, and parallel execution.
• Leadership experience as a technical lead driving design and delivery.
• Deep understanding of GPU architectures (SMs, Warp scheduling, Tensor Cores).
• Expert-level Triton or CUDA; contributions to Triton compiler or CUDA kernels are a plus.

Responsibilities

• Lead the technical strategy for benchmarking and performance optimizations at the inference engine and GPU kernel layers, maximizing TFLOP value.
• Engineer solutions for complex performance issues, including attention, memory and precision management, with multi-node GPU parallelization.
• Proactively implement cutting-edge optimization techniques to keep DigitalOcean at the forefront of Gen AI.
• Identify kernel fusion opportunities for GLM-5 kernels across Transformer blocks (FlashAttention, RMS Norm).
• Tune gateway router kernels for MoE models like Qwen3-235B, DeepSeek V3, GLM-5.
• Advise on hardware procurement and software integration.
• Develop state-of-the-art quantization techniques (FP8, INT8, FP4) to double throughput without accuracy loss.
• Lead by example through high-quality code and design reviews.
• Collaborate with Product Management to translate hardware limits into shippable features.
• Contribute to open-source AI communities.

Skills

Performance Architecture

Deep-Dive Optimization

Technological Innovation

Hardware Fluency

Precision Optimization

Technical Mentorship

Strategic Collaboration

Low-Level Mastery

Triton/CUDA

Tools

CUDA

ROCm

TensorRT

OpenAI Triton

Job description

DigitalOcean is seeking a Senior Engineer 2 to drive performance optimizations for AI inference across GPU architectures, focusing on memory bandwidth, throughput, and low-latency inference at scale.

You will lead benchmarking, optimize attention layers, and champion quantization techniques, while mentoring peers and aligning with product goals to deliver high-performance, developer-friendly AI infrastructure.