全职 【RD】AI System Validation Engineer 工作, 薪水, Acer Inc. 新北市, 台灣 公司招聘中 - Ricebowl

We are looking for an AI System Validation Engineer with experience in AI system testing and performance validation. This role will focus on the testing, benchmarking, and validation of AI server platforms, GPU systems, and AI solutions.
The position is dedicated to MLPerf testing, hardware benchmark testing, and AI solution testing, with the goal of producing reliable performance data and validation results to support internal product readiness, solution optimization, and customer-facing technical evaluation.
Key Responsibilities

• Execute MLPerf Training / Inference tests to validate the performance of AI server and GPU platforms.
• Plan and perform hardware benchmark testing, including server, GPU, storage, and overall system performance validation.
• Conduct AI solution testing, including functional validation, performance testing, stability testing, and deployment verification.
• Build and maintain test environments for AI workloads, such as LLM inference, model training, RAG, and agent-based AI solutions.
• Analyze test and benchmark results, identify system bottlenecks, and provide recommendations for performance and stability improvement.
• Prepare test plans, test cases, benchmark reports, and validation documentation.
• Develop test automation tools and scripts using Python / Shell Script to improve testing efficiency and result collection.
• Collaborate with R&D, product, and solution teams to improve platform compatibility, reliability, and delivery readiness.
• Support AI solution validation in Linux, container, and Kubernetes environments.

我們正在尋找具備 AI 系統測試與效能驗證經驗的工程師，負責 AI 伺服器平台、GPU 系統，以及 AI 解決方案的測試、Benchmark 與驗證工作。
此職位將聚焦於 MLPerf 測試、硬體效能測試（Hardware Benchmark Testing）以及 AI Solution Testing，產出具參考價值的測試數據與分析報告，協助內部產品驗證、方案優化與客戶技術支援。
主要工作項目

• 執行 MLPerf Training / Inference 測試，驗證 AI 伺服器與 GPU 平台之效能表現。
• 規劃並執行 硬體 Benchmark 測試，包含伺服器、GPU、儲存與系統整體效能驗證。
• 執行 AI 解決方案測試，包含功能驗證、效能測試、穩定性測試與部署驗證。
• 建立與維護 AI 工作負載測試環境，例如 LLM inference、model training、RAG/Agent-based AI solution 等。
• 分析測試結果，找出系統瓶頸，提出效能優化與穩定性改善建議。
• 撰寫測試計畫、測試案例、測試報告與 Benchmark 文件。
• 以 Python / Shell Script 開發測試自動化工具，提升測試效率與結果蒐集能力。
• 與研發、產品、解決方案團隊合作，提升 AI 平台與 AI 方案的相容性、可靠性與可交付性。
• 協助 Linux、Container、Kubernetes 環境下之 AI Solution 驗證

Required

• Solid hands-on experience with Linux
• Experience in AI / GPU server testing, performance validation, or benchmarking
• Familiarity with MLPerf Training / Inference concepts and testing flow
• Experience in hardware performance testing, including server / GPU / storage benchmark execution
• Familiarity with AI/ML frameworks such as PyTorch and TensorFlow
• Understanding of LLM deployment and validation requirements, including latency, throughput, and stability metrics
• Ability to use Python, Bash, or Shell Script for test automation and data processing
• Strong ability to analyze test results, identify bottlenecks, and propose optimization recommendations
• Good cross-functional communication and teamwork skills

Preferred

• Familiarity with CUDA, ROCm, oneAPI, or related AI/GPU toolchains
• Basic hands-on experience with Docker / Containers / Kubernetes
• Experience in testing LLM, RAG, Chatbot, or AI Agent solutions
• Experience in AI server, GPU platform, or HPC system validation projects
• Experience preparing benchmark reports and supporting internal or customer-facing technical discussions

必備條件

• 熟悉 Linux 作業系統與基本系統操作
• 具備 AI / GPU Server 測試、效能驗證或 Benchmark 相關經驗
• 熟悉 MLPerf Training / Inference 測試流程與觀念
• 具備 硬體效能測試 經驗，能執行伺服器 / GPU / 儲存相關 Benchmark
• 熟悉 AI/ML Framework，如 PyTorch、TensorFlow
• 理解 LLM 部署與測試需求，包含 latency、throughput、stability 等驗證指標
• 能使用 Python、Bash 或 Shell Script 進行測試自動化與資料整理
• 具備測試結果分析能力，能定位瓶頸並提出改善建議
• 具備跨部門合作與良好溝通能力

加分條件

• 熟悉 CUDA、ROCm、oneAPI 等 AI / GPU 工具鏈
• 熟悉 Docker / Container / Kubernetes 基本操作
• 具備 LLM、RAG、Chatbot、AI Agent 等 AI Solution 測試經驗
• 曾參與 AI Server、GPU Platform 或 HPC 系統之效能驗證專案
• 具備 Benchmark 報告撰寫與對內 / 對客戶技術說明經驗

Acer-RCM-Adapting and Coping

Acer-RCM-Analysing and Interpreting

Acer-RCM-Creating and Conceptualising

Acer-RCM-Enterprising and Performing

Acer-RCM-Interacting and Presenting

Acer-RCM-Leading and Deciding

Acer-RCM-Organising and Executing

Acer-RCM-Supporting and Co-operating