Senior AI Agent Architect: Multimodal Dialogue & Reasoning

Shopee; Singapore; On-site; SGD 120,000 - 180,000; Full time

Job summary

Shopee is seeking an experienced researcher-engineer for the Engineering and Technology team in Singapore. You will design and implement core agent algorithms, including multi-turn dialogue planning, tool orchestration, and memory architectures using knowledge graphs to support personalization and session persistence.

Work involves emotion-aware dialogue modeling, LLM alignment and safety, multimodal reasoning, and deployment across GPU clusters.

Qualifications

• Hands-on research and engineering in conversational agents.
• Experience with knowledge-graph memory and persona-aware dialogue.
• Publications in top AI/NLP venues desired.
• Proficiency in Python, Bash; PyTorch and Hugging Face.

Responsibilities

• Design and develop core agent algorithms, including multi-turn dialogue planning.
• Build agent memory architectures using knowledge graphs.
• Develop emotion-aware dialogue modeling techniques.
• Design LLM alignment and safety strategies (SFT, DPO).
• Benchmark and deploy LLMs/VLMs on GPU clusters and cloud environments.

Skills

Python

PyTorch

HuggingFace

Knowledge graphs

Reinforcement learning

NLP

Research

Bash

Education

Master's degree in CS

Tools

AWS EC2

PostgreSQL

Neo4j

AWS S3

Job description

Shopee is seeking an experienced researcher-engineer for the Engineering and Technology team in Singapore. You will design and implement core agent algorithms, including multi-turn dialogue planning, tool orchestration, and memory architectures using knowledge graphs to support personalization and session persistence.

Work involves emotion-aware dialogue modeling, LLM alignment and safety, multimodal reasoning, and deployment across GPU clusters.