Overview
We are seeking an AI Researcher to conduct research and development in Agentic AI, LLM Reasoning, Information Retrieval, Memory Systems, and Personalized AI Agents. The role focuses on designing, evaluating, and improving next-generation AI systems that combine retrieval, reasoning, memory, planning, and tool usage to solve complex real-world tasks.
Responsibilities
- Research and develop agentic AI systems capable of multi-step reasoning, planning, and decision-making.
- Design and evaluate retrieval-augmented systems (RAG), semantic search, memory architectures, and knowledge-based reasoning frameworks.
- Develop and benchmark multi-agent systems, tool-using agents, and autonomous workflows.
- Conduct scientific experimentation, ablation studies, benchmarking, and failure analysis on LLM systems.
- Research long-term memory, personalization, and adaptive learning mechanisms for AI agents.
- Explore synthetic data generation, fine-tuning, and alignment techniques to improve LLM capabilities.
- Publish research findings and translate research prototypes into deployable AI solutions.
Technical Skills
- Strong understanding of LLMs, NLP, Generative AI, and Agentic AI.
- Experience with reasoning systems, planning frameworks, tool calling, or multi-agent architectures.
- Hands-on experience with RAG, embeddings, vector databases, semantic retrieval, GraphRAG, or knowledge graphs.
- Experience evaluating AI systems through benchmarking, experimentation, and ablation studies.
- Familiarity with LoRA, QLoRA, PEFT, SFT, DPO, RLHF, or model alignment techniques.
- Experience with synthetic data generation and AI evaluation frameworks.
- Strong Python skills and familiarity with PyTorch, Hugging Face, LangGraph, AutoGen, CrewAI, or similar frameworks.
Singapore citizen / PR only due to quota restriction.
