Staff Engineer, AI Inference Optimization

DigitalOcean; Boston (MA); On-site; USD 191,200 - 239,000; Full time

Benefits offered by this job

Equity compensation

Bonus potential

Conference reimbursement

LinkedIn Learning access

Flexible time off

Job summary

DigitalOcean is seeking a Senior Engineer 2 to play a key technical role in our AI Inference Optimization team. Responsible for architectural decisions maximizing throughput and minimizing latency for large models, you will lead performance strategies across GPU clusters and guide the technical roadmap.

You will mentor through code reviews, collaborate with product teams, and push cutting-edge optimization techniques to keep DigitalOcean at the forefront of Gen AI performance.

Qualifications

• 5+ years in high-performance computing or AI infrastructure.
• Deep familiarity with Gen AI landscape and model families.
• Hands-on optimization of attention layers and distributed GPUs.
• Understanding of NVIDIA/AMD GPUs and software stacks.
• Experience integrating and contributing to open-source projects.
• Strong system design for low-level GPU programming and memory access.
• Technical lead experience guiding design and delivery.

Responsibilities

• Lead benchmarking and performance optimization at engine and kernel layers.
• Engineer solutions for attention, memory, and FP8/BF16 tradeoffs.
• Identify kernel fusion opportunities and optimize across multi-node GPUs.
• Advise on hardware procurement and software integration.
• Mentor through code reviews and drive cross-functional alignment.
• Collaborate with product managers to translate hardware limits into features.

Skills

Technical Depth

Gen AI Literacy

Optimization Expert

Hardware Fluency

Open Source Mastery

Systems Design

Leadership through Influence

Low-Level Mastery

CUDA/Triton

Tools

CUDA

ROCm

TensorRT

OpenAI Triton

Triton compiler

Job description

DigitalOcean is seeking a Senior Engineer 2 to play a key technical role in our AI Inference Optimization team. Responsible for architectural decisions maximizing throughput and minimizing latency for large models, you will lead performance strategies across GPU clusters and guide the technical roadmap.

You will mentor through code reviews, collaborate with product teams, and push cutting-edge optimization techniques to keep DigitalOcean at the forefront of Gen AI performance.