Tai Tiet

About Tai Tiet

Senior Systems Engineering with 10+ years of experience driving large-scale deployment, validation, and optimization of next-generation Telco, Medical System, and AI accelerator platforms. Proven track record leading cross-functional initiatives that bring advanced AI servers, GPU platforms, and datacenter-scale solutions from early concept to customer-ready production.Extensive hands-on experience with Intel Max, Flex, and Gaudi 2/3 accelerators, as well as NVIDIA A100, H100, Blackwell B200, and RTX Pro Blackwell platforms. Leadership responsibilities have included system integration, firmware and BIOS readiness, distributed training performance tuning, and enterprise-level workload validation in collaboration with hardware, firmware, and software teams.At Supermicro, led the development of flagship AI and multi-agent demos showcased at global events such as NVIDIA GTC, Cloud Fest, and SC—demonstrating technical vision, execution oversight, and the ability to translate emerging AI technologies into impactful customer-facing solutions. Projects included multi-agent Slack automation systems, real-time LLM interaction pipelines, and generative AI workflows running on Gaudi and Blackwell platforms.Strong background orchestrating AI cluster deployments across datacenters, including networking architecture (InfiniBand/RoCE), PXE-based provisioning pipelines, GPU/accelerator integration, Kubernetes-based orchestration, and full-stack system reliability testing. Frequently serve as a technical partner to enterprise customers, supporting PoCs, performance optimization, and best practices for deploying AI at scale.With a foundation in Computer Engineering and a Master’s degree in Data Science, work naturally bridges hardware engineering, AI/ML pipelines, and customer solution delivery. Passionate about team leadership, mentoring engineers, and shaping technical roadmaps that advance organizational goals and accelerate adoption of cutting-edge AI technologies.Actively pursuing opportunities to step into a Staff/Senior Engineering role, guiding teams and enabling organizations to build the next generation of high-performance AI and datacenter systems.

Experience

1. Engineer

   美超微

   Mar 2023 — Present · San Jose, CA, US

   Lead engineer for enterprise AI demo development used at major industry events including NVIDIA GTC, SuperCloud, SC, and Computex, accelerating adoption of Supermicro’s latest AI and HPC platforms.Designed, built, and validated end-to-end AI systems using the full Intel AI accelerator portfolio, including Intel Max Series GPUs, Intel Flex Series, and Intel Gaudi 2/3, ensuring performance, stability, and customer-ready reliability.Extensive experience deploying and optimizing NVIDIA platforms at scale, including A100, H100, B200 Blackwell, and RTX Pro Blackwell GPUs across multi-node high-density Supermicro servers.Deep familiarity with NVIDIA AI Enterprise (NVAI-E), NVIDIA Blueprint architectures, NGC containers, Triton Inference Server, vLLM, NIM microservices, and Multi-Instance GPU (MIG).Built full-stack AI demo pipelines integrating: LLM, Speech models, Vision & generative AI models, Multi-Agent frameworks. Created a flagship demo of two AI models conversing on Slack, showcasing multi-agent communication, event-driven automation, and real-time generative reasoning for enterprise applications.Built a Multi-Agent Slack Assistant capable of retrieval, reasoning, coding, translation, summarization, and image generation—running seamlessly on Intel Gaudi 3 or NVIDIA Blackwell platforms.Provided technical support and solutions engineering for customers during trial evaluations, PoC testing, and datacenter deployment—troubleshooting:GPU/accelerator hardware, firmware/BIOS/BMC, networking (RoCE, InfiniBand, 100/200/400/800GbE), storage and cluster orchestration.Supported deployment of large AI clusters including multi-node GPU configurations, Kubernetes clusters, and high-speed network fabrics for training/inference workloads.Collaborated with Intel, NVIDIA, and Supermicro product teams to qualify, certify, validate next-generation accelerators and optimize performance before public release.

Education

• UC Irvine

  Bachelor of Science , Computer Engineering

• University of Washington

  Master of Science , Information Management with specialize in Data Science