AI Application Engineer (Regional)

Our client is hiring an AI Application Engineer with hands-on experience in delivering real-world LLM-powered systems. This role focuses on engineering, system design, and production deployment of AI applications rather than research-only work. 
You will own or co-own the full lifecycle of AI projects, including model selection, RAG architecture, multi-agent orchestration, prompt workflows, and system optimization, translating cutting-edge AI capabilities into scalable business solutions. 

This is a global hiring role. We welcome candidates worldwide with proven AI project delivery experience. 

AI Application Development & Delivery 

• Lead or co-own AI projects from requirement analysis, architecture design, to production deployment. 
• Fine-tune, align, and optimize general LLMs, embedding models, and inference pipelines. 
• Design structured and reusable Prompt Engineering workflows to improve task reliability and performance. 

Multi-Agent Systems & Reasoning Frameworks 

• Build and orchestrate multi-agent workflows using LangChain, LangGraph, MCP, or similar frameworks. 
• Implement reasoning paradigms such as ReAct, Chain-of-Thought (CoT), Tree-of-Thought (ToT) to enhance agent decision quality and controllability. 
• Integrate or evaluate agent platforms such as Coze, FastGPT, Dify when applicable. 

RAG & Knowledge Systems 

• Design and implement Retrieval-Augmented Generation (RAG) architectures. 
• Build document knowledge retrieval systems using vector databases (Milvus, FAISS, Chroma). 
• Continuously improve retrieval quality, context relevance, and reasoning accuracy. 

Audio & Voice Interaction 

• Design or integrate audio-based interaction capabilities, including speech input, transcription, and voice output. 
• Design agent interaction flows for audio-based scenarios. 
• Optimize latency, reliability, and user experience for real-time or near-real-time voice interactions. 

Technical Research & Knowledge Sharing 

• Track emerging AI trends (model alignment, multi-agent systems, multimodal models). 
• Contribute to internal documentation, benchmarks, prototypes, or technical sharing. 

Qualifications

• Bachelor’s degree or above in Computer Science, AI, or a related field. 
• Proven experience delivering at least one end-to-end AI application (LLM / RAG / Agent system). 
• Hands-on experience with Prompt Engineering, LangChain, or RAG frameworks. 
• Experience designing multi-agent architectures, preferably with LangGraph or MCP. 
• Familiarity with vector database selection and integration. 
• Strong understanding of ReAct-style agent workflows. 

Nice to Have 

• Practical experience with LoRA / QLoRA, model alignment, or inference optimization. 
• Familiarity with streaming or real-time speech processing architectures. 
• Experience in model training or reinforcement learning fine-tuning. 
• Open-source contributions, technical blogs, or internal/external tech sharing. 
• Strong cross-functional communication skills.