22 hours ago
Seoul, Korea, SouthEntry Level
Responsibilities
- Build routing intelligence for vllm-sr, including semantic, policy-based, workload-aware, safety, privacy, tools, and agent routing.
- Develop a production-ready serving engine for decision models and integrate it with vLLM, Kubernetes, Envoy, Gateway API, and AI gateways.
- Optimize model inference across batching, memory, latency, throughput, and GPU kernels using HIP or Triton on ROCm.
- Train, post-train, evaluate, and improve foundation decision models and automate the associated research pipeline.
- Co-design decision models and the serving engine, build benchmarks, and measure routing quality, latency, and cost.
- Ship production-quality software with testing, continuous integration, observability, and reliability practices.
- Review and land open-source contributions, triage issues, ship releases, and help grow the vllm-sr community.
- Represent AMD and vllm-sr through conferences, workshops, talks, demos, blogs, and documentation.
- Collaborate with ROCm, kernel, AI framework, Enterprise, and IT teams to validate the system on AMD GPU platforms and support internal products and services.
Requirements
- Fresh graduates with strong Python programming skills and proficiency in at least one of Go, Rust, or C++.
- Hands-on experience serving, training, or building products with large language models, such as through vLLM or SGLang.
- Daily use of AI coding agents and agentic workflows, with the ability to verify and correct generated output.
- Proven end-to-end delivery of AI-driven products from ideation and design through deployment and operations.
- Strong fundamentals in systems, machine learning, or both.
- Experience with GitHub workflows, public code review, and clear written English.
- Open-source contributions to projects such as vllm-sr, vLLM, SGLang, PyTorch, Hugging Face, Envoy, or Kubernetes.
- GPU programming experience with HIP, CUDA, or Triton, or experience with ROCm on AMD Instinct GPUs.
- Experience with model routing, API gateways, or LLM inference optimization.
- Experience training or post-training classifiers, reward models, or multimodal models.
- Papers at top machine learning or systems conferences are valued.
- Technical writing, public speaking, or community-building experience is valued.
- Bachelor's or Master's degree in Computer Science, Computer Engineering, Electrical Engineering, or a related field, or equivalent experience.
- Research experience in machine learning systems, distributed systems, or high-performance computing is a plus.
Benefits
- Role based in Seoul, Korea; AMD benefits are referenced but not detailed in the posting.
Categories
About AMD
AMD designs and sells CPUs, GPUs, and adaptive/embedded computing products for PCs, data centers, gaming, and edge devices. Its portfolio includes Ryzen and EPYC processors, Radeon and Instinct graphics, and adaptive SoCs from its Xilinx acquisition, sold to OEMs, cloud providers, and device makers. Founded in 1969 and headquartered in Santa Clara, it is a public company on NASDAQ and supplies semi-custom chips for major game consoles.