Responsibilities

● Work with business units to identify AI use cases and gather requirements for LLM-powered applications.

● Design and deploy AI platform components — LLM Gateway (routing, API abstraction, quota & observability), RAG pipelines, prompt/agent orchestration, and AI application frameworks; evaluate alternatives and coordinate proofs of concept.

● Track the AI ecosystem and technology trends, and turn that insight into architecture and product decisions.

● Apply the AI platform to real-world use cases (e.g. enterprise search, document Q&A, code generation, AI-assisted forecasting).

● Ensure responsible AI: data security, model governance, cost control, and output quality.

● Build up the team's AI knowledge and skills.

Required Knowledge/Skills, Education, and Experience

● Master's degree in Computer Science, Computer Engineering, Electrical Engineering, Mathematics, or Physics.

● Deep understanding of the AI ecosystem and its direction — commercial and open-source models, major AI frameworks and toolchains — and the ability to make sound technology and architecture decisions based on it.

● Substantial hands-on experience with AI platforms and technologies: designing, building, and operating production AI systems, including LLM Gateway patterns, end-to-end RAG delivery (chunking, embedding, vector retrieval, reranking), and production AI application development.

● Familiarity with modern AI architecture: LLM APIs, gateway/proxy patterns, and AI application frameworks; experience with at least one LLM ecosystem (OpenAI, Anthropic, Azure OpenAI, or open-source models via vLLM/TGI) and frameworks such as LangChain or LlamaIndex.

● Experience with loosely coupled software architecture involving APIs and secure data transfer.

● Strong skills in designing, prototyping, implementing, and refactoring software; compiling and maintaining technical documentation and presentations for non-technical audiences.

● Good communication and collaboration skills with both developers and non-developers; able to work independently and proactively. Excellent written and verbal English.

Preferred Knowledge/Skills, Education, and Experience

● Knowledge of industrial automation, production, or engineering.

● Experience with vector databases (Milvus, Pinecone, Elasticsearch vector search, pgvector), model serving (vLLM, Triton), and LLMOps/observability tooling.

● Knowledge of Spark, Flink, Kafka; SQL.

● Knowledge of AWS, Azure, Snowflake, and Alibaba Cloud (Model Studio/百炼, PAI, Qwen/通义千问, OSS, ACK, Function Compute).

● Familiarity with fine-tuning, quantization, or LLM evaluation frameworks.

● Prior experience building enterprise AI gateways or internal AI developer platforms.

● Active engagement with the AI community (open-source contributions, blogs, talks).