AI Data Engineer

[About the Team and Our Impact]
We solve real-world retail challenges through deep R&D and practical technology implementation.
Our recent work has earned prominent technical recognition in Taiwan and internationally:

■ Selected for NVIDIA GTC Poster presentations in 2025 and 2026: Our technical achievements were invited for presentation at NVIDIA's annual global technology conference (GTC).
■ Selected for the 2025 IT Matters Awards (AI Selected): Named an AI Application Project of the Year, recognizing the team's technical innovation and real-world commercial impact.
■ Expanding across domestic and international markets: Our in-house AI retail solutions have been deployed in Taiwan and overseas and are moving toward large-scale commercial rollout across multiple countries.

[Role Overview]
We develop smart-retail applications for unmanned stores, smart retail cabinets, and automated vending systems. Join our core R&D team to turn AI technology into tangible business value.

[Key Responsibilities]
1. LLM Engineering & Domain Adaptation
•Model Training & Finetuning: Execute LLM Pre-training, Continual Pre-training, and Finetuning tailored to specific domain knowledge.
•Data Pipeline Development: Utilize Large Language Models (LLMs) for complex data processing tasks, including data cleaning, field alignment, classification, and data augmentation.
•Search & Retrieval: Design and build Retrieval-Augmented Generation (RAG) pipelines and integrate vector databases (such as FAISS, Qdrant, Weaviate, Milvus, and Pinecone).
•Advanced Search Implementation: Implement vector, semantic, and multimodal search technologies to enhance intelligent information retrieval and data comprehension.

2. Distributed Training & Cluster Management
•Infrastructure Design: Design, build, and maintain Kubernetes (K8s) and SLURM clusters optimized for intensive AI applications.
•Distributed Training: Configure and manage Distributed Data Parallel (DDP) training across multi-node and multi-cluster environments.
•Container Management: Manage and optimize container images and deployments across multiple Google Cloud Platform (GCP) environments to ensure reliable scaling.

3. Inference Optimization & Serving
•High-Performance Serving: Deploy and optimize LLMs using advanced inference engines and frameworks, including Triton Inference Server, vLLM, TensorRT-LLM, and SGLang.
•Resource Efficiency: Ensure low-latency and high-throughput model delivery while balancing hardware and GPU utilization.

4. Cloud Infrastructure & Resource Management
•Multi-Cloud Management: Oversee compute and AI resources across GCP, AWS, and Azure OpenAI environments.
•Security & Compliance: Manage Identity and Access Management (IAM) policies, roles, and cloud security measures.
•Infrastructure as Code (IaC): Automate resource provisioning and infrastructure management using tools such as Terraform or CloudFormation.
•Cost Optimization: Monitor, analyze, and optimize cloud resource consumption to manage operational costs.

5. Integration, Message Queues, & Monitoring
•Queue Services: Integrate and host reliable messaging and notification systems, such as Amazon SQS, Google Cloud Pub/Sub, and RabbitMQ.
•Observability & APM: Implement and manage Application Performance Monitoring (APM) systems (e.g., Datadog, Prometheus) and distributed tracing tools (e.g., Jaeger, Zipkin) to monitor application uptime, latency, and performance.