Solution Architect - Agentic AI - CSP
Solution Architect at NVIDIA focused on designing and optimizing large-scale LLM and Agentic AI inference systems for cloud service providers, leveraging NVIDIA accelerated computing platforms and tools like TensorRT-LLM.
- Compensation
- Not disclosed
- Location
- China, Beijing
- Work setup
- Not specified
- Seniority
- Not specified
- Employment
- Not specified
- Experience
- 3+ years
- Category
- Architecture
- Posted
Skills
Must have
- MS or PhD in Electrical Engineering, Computer Science/Engineering, Mathematics, Physics, or a related field, or equivalent practical experience
- 3+ years of experience in large language models, AI inference, distributed AI systems, or related areas
- Hands-on experience in system design, performance optimization, or large-scale AI workloads
- Strong understanding of modern AI systems and ability to analyze performance across models, frameworks, compute, memory, communication, and infrastructure
- Strong programming and problem-solving skills with experience using common AI frameworks and tools
- Clear and structured thinking, strong technical communication skills, and ability to work with engineering teams and senior stakeholders
Nice to have
- Experience with Agentic AI or advanced LLM inference optimization for large-scale production workloads
- Demonstrated experience optimizing AI workloads on GPU-accelerated systems or distributed computing platforms
- Familiarity with NVIDIA Dynamo, TensorRT-LLM, disaggregated serving, heterogeneous inference, expert parallelism, or large-scale serving techniques
- Experience with NVIDIA AI software and platforms or with large cloud service providers and production AI infrastructure
- Proficiency in using AI tools or agents to improve engineering productivity
Full description
NVIDIA’s Solution Architect team is looking for an AI-focused Solutions Architect specializing in Agentic AI inference. We work with the most advanced computing hardware and software, driving the latest breakthroughs in artificial intelligence. As AI applications rapidly evolve from traditional model serving toward more complex agentic systems, inference infrastructure and optimization are becoming increasingly important to delivering scalable, responsive, and cost-efficient AI services.
We are looking for someone with strong expertise in large language models, AI inference, or distributed AI systems, and a passion for solving emerging challenges in Agentic AI. You will work closely with leading cloud service providers and AI platform teams to design and optimize large-scale inference solutions on NVIDIA accelerated computing platforms. You will also collaborate across NVIDIA Engineering, Product, Sales, and Business Development teams to turn customer requirements into scalable solutions and product improvements.
As a Solutions Architect, you will be the first line of technical expertise between NVIDIA and our customers. Your work will range from architecture design and proof-of-concept development to hands-on performance analysis and optimization of production-scale AI systems. Dynamically engaging with developers, AI researchers, infrastructure engineers, platform architects, and senior technical leaders is a meaningful part of the role.
What you’ll be doing:
- Work with customers to design, deploy, and optimize large-scale LLM and Agentic AI inference systems on NVIDIA accelerated computing platforms.
- Analyze end-to-end inference workloads and identify opportunities to improve latency, throughput, scalability, resource utilization, and cost efficiency.
- Explore and develop solutions for emerging Agentic AI inference challenges, which may include areas such as KV cache management, intelligent routing and scheduling, distributed or disaggregated serving, heterogeneous inference, and large-scale model parallelism.
- Build proof-of-concepts, performance studies, and reference solutions to validate new architectures and demonstrate the value of NVIDIA technologies in real-world AI workloads.
- Partner with NVIDIA Engineering, Product, Sales, and Business Development teams to solve customer problems, provide product feedback, and help shape technical strategies for strategic AI infrastructure opportunities.
##
What we need to see:
- MS or PhD in Electrical Engineering, Computer Science/Engineering, Mathematics, Physics, or a related field, or equivalent practical experience.
- 3+ years of relevant experience in , with hands-on experience in system design, performance optimization, or large-scale AI workloads.