About our client A fast-growing technology group is building a new AI department from the ground up in Singapore, backed by large-scale multilingual data and significant R&D resources. The team operates with startup agility - focused on shipping, iterating, and solving real problems at scale. What you'll work on Design and implement core Agent architectures: task planning, state tracking, multi-turn conversation management Build advanced RAG systems with hybrid retrieval (BM25 + Embedding) and LLM-driven proactive retrieval Develop long/short-term memory systems and tool/plugin dispatching pipelines Own end-to-end delivery - intent recognition accuracy, task completion rates, and latency optimisation in production What we're looking for Master's in CS or NLP strong understanding of Agent frameworks (LangChain, LlamaIndex, ADK) Hands-on experience with RAG, vector retrieval, and Reranker model optimisation Solid Prompt Engineering skills familiarity with LLM capability boundaries Plus: ...