About The Team

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 Intelligent Customer Service Chatbot team at Shopee Singapore, building multilingual dialogue systems that serve consumers and merchants at scale. We apply LLMs, Agents, RL, and modern agent engineering (context engineering, agent harness design, agentic loops) to real-world customer service.

We have deployed LLM-powered CS systems across multiple markets with significant gains in service quality and efficiency. We're now building a unified human-AI collaboration system where human feedback drives continuous iteration, evolving our chatbot from a dialogue system into an intelligent problem-resolving agent capable of complex task execution — contextual understanding, autonomous planning, tool use, and self-evolution.

Job Description

You will help design and develop core algorithms and agent systems, including:

Context engineering — retrieval, compression, and injection of multi-source context (profile, memory, history, tool results)
Agent harness design — tool orchestration, control flow, error recovery, and sub-agent coordination
Agentic loops — design and convergence of plan–act–observe loops for long-horizon tasks
Memory architecture — layered memory construction and read/write strategies
Prompt engineering — AI-friendly skills/SOPs and automated bad-case patching
SFT & alignment — SFT distillation and RL alignment (including reward design)
RAG & tool integration — resolution-oriented retrieval and MCP/tool integration
Eval & observability — high-discrimination AutoQA and trace-based iteration

You'll work across product, business, operations, and engineering through the full lifecycle — from problem definition to deployment and iteration.

Requirements

Current pursuing a Bachelor's in CS, CE, or related field; Master's or above in CS, computational linguistics, or related majors preferred
Excellent coding skills with proficiency in at least one of C++/Golang/Python/Java
Strong foundation in algorithms and data structures
Familiar with at least one of: NLP, LLMs, RL, Agent systems, or Multimodality
Experience with a deep learning framework (e.g. PyTorch, TensorFlow)
Familiarity with agentic engineering (tool use, context management, agent eval) and tooling like MCP / Langfuse is a plus
Good communication and collaboration skills
Full-time interns (6 months) preferred; part-time (≥3 days/week) also welcome

Algorithm Engineer Intern, Marketplace Intelligence & Data - Chatbot