About The Team
Shopee will be prioritizing applicants who have a current right to work in Singapore, and do not require Shopee sponsorship of a visa.
Kindly note that you can only be considered in one recruitment process at a time within Sea Group and will be considered for jobs in the order that you have applied.
The Marketplace Intelligence and Data team's mission is to build sustainable and efficient data and intelligence products to facilitate Shopee's business development. The team is responsible for Shopee e-commerce data warehouse construction, merchant and operation data product construction, all-link traffic data, product algorithms, including product release, control, information optimization, SPU library and its comparison business, marketing algorithms, including Merchandising, Product Selection, Recommendation Algorithm, Evaluation Algorithms, User Profiling, and in addition, basic AI capabilities, such as Machine Translation, Speech Algorithm, Image Algorithm, and Real-person Authentication.
We are the Shopee Intelligent Customer Service Chatbot team based in Singapore, dedicated to exploring cutting-edge AI technologies, including but not limited to large language models (LLMs), natural language processing, deep learning, and knowledge graphs. Our mission is to tackle core challenges in multilingualism, context, and human-machine collaboration in Q&A and dialogue systems. We aim to revolutionize the service experience for consumers in Southeast Asia and globally.
By integrating state-of-the-art algorithmic technology, robust engineering foundations, thoughtful product design, and intelligent operations, our team has developed a rich, efficient, and high-performance Q&A technology system. This system significantly supports markets in Southeast Asia, including Singapore and Indonesia, as well as the South American market, including Brazil, effectively addressing the vast differences in customer service demands and controlling costs while improving customer satisfaction.
The rise of large language language models has opened up vast possibilities in the field of customer service. Natural dialogue and task-driven scenarios are crucial stages for showcasing the capabilities of LLMs. Here, you have the opportunity to build algorithmic systems based on large language models, redefining chatbots and agent assistance products, and bringing a novel user experience to a broad consumer base.
Job Description
- Design and develop algorithms for AI assistants and agentic applications, covering areas including LLM post-training, supervised fine-tuning (SFT), Mixture-of-Experts (MoE), and user preference alignment using reinforcement learning or preference optimization techniques such as DPO and GRPO.
- Build LLM-powered recommendation systems for user intent understanding and recommendation, including retrieval, coarse and fine ranking, user sequential modelling, semantic representation learning, and generative recommendation.
- Develop production-ready LLM applications that integrate models with tools, APIs, knowledge bases, memory systems, and business workflows to solve real-world customer service and recommendation problems.
- Explore and implement advanced agent harness engineering techniques, including:
- Agent loop and graph orchestration
- Planning, reasoning, tool selection, and execution
- Multi-agent collaboration and task delegation
- Skill design, skill mining, and skill lifecycle management
- Context, memory, and knowledge management
- Reflection, feedback loops, and self-improvement mechanisms
- Evaluation, observability, failure analysis, and guardrail design
- Design and develop reusable algorithm platforms, evaluation frameworks, experimentation pipelines, and supporting tools to improve the development, deployment, and iteration efficiency of AI assistant capabilities.
- Establish robust offline and online evaluation methodologies for AI assistants and agents, covering task success rate, response quality, tool-use accuracy, safety, latency, cost efficiency, user satisfaction, and business impact.
- Collaborate closely with product, business, operations, data, and engineering teams throughout the complete lifecycle of intelligent customer service products—from problem definition and solution design to development, deployment, performance monitoring, root-cause analysis, and continuous user-experience improvement.
- Track emerging developments in LLMs, agents, recommendation systems, and AI engineering, and translate relevant research or industry best practices into practical solutions for production environments.
Requirements
- Bachelor's degree or higher in Computer Science or a related field
- Proficiency in at least one programming language, such as Python, C++ & Go
- Skilled in at least one mainstream deep learning framework, like PyTorch, TensorFlow
- Passion for LLM-related technology and solving challenging real-world problems
- Deep understanding of computer fundamentals, including data structures, algorithms, and machine learning
- Familiarity with mainstream algorithmic models in large language models, natural language processing, recommendation, and multimodality (including principles and implementation)
- Prior research or internship experience in natural language Q&A or recommendation technology is preferred
