AMD

2 weeks ago

Principal AI Performance Engineer - LLM Inference (SGLang)

✓ Verified Job

Canonical indexed version, validated from employer's careers page.

Source: Company careers pageValidated by: Fursa AI

Last checked: 8 hours agoStatus: Live

Job Summary

Lead end-to-end AI inference performance optimization on AMD GPUs using SGLang. Work with a small, highly technical team to profile, diagnose, and optimize cross-stack bottlenecks—from GPU kernels through SGLang scheduling and RadixAttention/prefix caching—across customer-relevant serving configurations. Deliver measurable uplifts, integrate custom kernels, advance shared methodology, and contribute upstream improvements to SGLang and related open-source frameworks.

Location: Helsinki, Stockholm, Cambridge

Workplace: Hybrid

Employment Type: Full time

Job Function: Data Science & Machine Learning

Seniority: Mid level

Key Responsibilities

•Drive end-to-end performance optimization on SGLang across leading models and customer-relevant serving configurations.
•Profile, diagnose, and resolve cross-stack performance bottlenecks across GPU kernels, operator dispatch, SGLang scheduler, RadixAttention/prefix caching, and multi-node communication.
•Diagnose kernel-level performance issues using profiling tools (occupancy limits, L2 cache thrashing, register pressure, memory coalescing) and turn findings into actionable optimizations.
•Lead customer-facing technical engagements by presenting results, recommending optimizations, and delivering measurable performance uplifts.
•Integrate and optimize custom kernels in SGLang and optimize multi-node distributed inference (communication-compute overlap, parallelism strategies, scale-out performance).

Pay and Benefits

Salary: EUR 99,050 - 141,500 annually

Key Requirements

•7+ years of software development experience in GPU computing, AI systems, or high-performance computing.
•Deep, hands-on experience with SGLang internals, including profiling, bottleneck diagnosis, and tracing from user requests to GPU kernels.
•Strong understanding of GPU kernel performance (occupancy, cache behavior, memory coalescing, instruction-level bottlenecks) and translating profiling data into optimizations.
•Experience with custom kernel development or integration (HIP, CUDA, Triton, or CK).
•Master’s or PhD in Computer Science, Computer Engineering, Electrical Engineering, or equivalent (advanced degree preferred).

Experience:7+ yearsGPU computingAI systemsHigh-performance computingDistributed inferenceOpen source

Education:Master's in Computer Science, Computer Engineering, Electrical Engineering (or equivalent)

Skills:Technical leadershipWritten and verbal communicationProblem-solvingPerformance optimization mindsetCustomer-facing presentation

Languages:English

Tech Stack:SGLangRadixAttentionTritonGluonCKPyDSLASMAITERVLLMTensorRT-LLMPyTorchPythonC++HIPCUDALinuxRCCLNCCLRDMATP

Company Brief

AMD

Designs and produces semiconductor products including CPUs, GPUs, and adaptive SoCs for consumer, enterprise, and embedded markets, competing across PCs, data centers, and gaming industries.

Industry: Electronics Manufacturing

Company Size: Enterprise (1,001+ employees)

Revenue: USD 1B+

Growth: Public Company

Valuation: Public Company (Market Cap in USD)

Funding: IPO / Publicly Listed

Headquarters: Santa Clara, United States

Founded: 1969

Glassdoor: 3.9

Website LinkedIn

Loading job details...

Preparing the role view and application actions.

AMD