【Campus Hire】Agent / LLM Algorithm Engineer – rednote（Trust&Safety）

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

Job summary

rednote is expanding its International Trust & Safety Algorithm Team to build a global AI-native content governance and understanding system. You will drive end-to-end R&D for AI Agents, LLM/VLM training, and multimodal content understanding, collaborating across cross-functional teams to maintain technical leadership.

A strong track record in research and publications is a plus. This role focuses on state-of-the-art agent architectures, safety, and governance for a global community, with

Qualifications

• Bachelor's degree or above in Computer Science, Artificial Intelligence, Machine Learning, Big Data, Mathematics, or a related field.
• In-depth understanding of large models and AI Agents; familiarity with LLM paradigms, prompt engineering (PE), retrieval-augmented generation (RAG), and evaluation frameworks; knowledge of Multi-Agent architectures (Multi-Agent, Context Engineering/Management, ReAct/PlanAct/CodeAct).
• Familiar with Agentic RL and related infrastructure; ability to apply SFT and RL to optimize models for specific problems.
• Research-oriented mindset; publications at top conferences are a strong plus.

Responsibilities

• Build on AI capabilities such as VLM/Agent to fuse multi-dimensional signals—text, images, video, and audio across content types and scenarios including notes, comments, accounts, transactions, and livestreaming.
• Establish and evolve Rednote's content feature and grading systems, feeding core pipelines in governance, search, and traffic control.
• Own end-to-end R&D of Agent systems for governance, content understanding, and editing, covering ReAct/Tool Use/Multi-Agent architectures and related optimization.
• Advance full-pipeline training of LLM/VLM, including pre-training, SFT, and RL, to build foundational multimodal AI capabilities.
• Continuously track advancements in the AI Agent domain and promote cutting-edge techniques, publishing papers at top conferences.

Skills

LLMs & AI Agents

Multi-Agent Systems

Prompt Engineering

ReAct/PlanAct/CodeAct

MCP/A2A/Function Calling

Education

Bachelor’s degree or above in CS/AI/ML/Big Data/Math

Tools

CodeRL

Reward Systems

SFT/RL Training

Job description

The International Trust & Safety (TnS) Algorithm Team is the safety foundation for rednote’s global expansion. We build content governance and understanding capabilities for a global community, spanning risk prevention, intelligent moderation, content understanding and creation. Built on our proprietary Zeus family of AI agents and VLM, we bring cutting-edge algorithms into real-world applications at scale, laying a strong foundation for rednote worldwide.

Globalization opens new frontiers in community governance and content understanding. Cross-language and cross-cultural semantic understanding, the relationship between culturally diverse content creation and authentic expression, evolving regulations and compliance requirements across countries, and emerging risks in the AI era all call for new technical solutions. Here, we are building an AI-native content governance and understanding system for the world from the ground up, with our proprietary agents and large models powering the full lifecycle of risk detection, policy generation, and recursive self-improvement. Our goal is to build AI that understands every language and culture, empowering people to express themselves authentically and safely.

If you are passionate about using cutting-edge algorithms to solve real-world challenges, join us in building a community where people around the world feel confident expressing themselves and can trust the space they share.

Job Responsibilities

Build on AI capabilities such as VLM/Agent to fuse multi-dimensional signals—text, images, video, and audio—from image-text and video notes, covering international content risk identification and enforcement across content types and scenarios including notes, comments, accounts, transactions, and livestreaming. Continuously strengthen proactive risk detection to optimize the trade-off among risk, experience, and efficiency.

2. Content Features & Grading

Establish and continuously evolve Rednote's content feature system and content grading system, with outputs feeding into core pipelines such as enforcement in governance, as well as search, recommendation, and traffic control—building a long-term competitive moat for content quality and safety across Rednote's global ecosystem.

3. Agent System R&D

Own end-to-end R&D of Agent systems for core scenarios including Rednote governance, content understanding, and editing, covering ReAct / Tool Use / Multi-Agent architectures, intent understanding, and task planning. Play a key role in Agent performance optimization—from tool design and context management to orchestration strategies and Agentic RL post-training—delivering state-of-the-art products.

Play a key role in full-pipeline training of LLM/VLM, including pre-training, SFT, and RL. Contribute to building foundational AI capabilities in multimodal content understanding, editing, and generation, driving the full loop from AI technology to business value.

5. Frontier Technology Exploration

Continuously track the latest advancements in the AI Agent domain, introduce and validate the feasibility of emerging technologies, document and share best practices, and promote the application of cutting-edge technologies in Rednote's business—maintaining the team's technical leadership and producing papers at top international conferences (CVPR/NeurIPS/ICLR/ACL, etc.).

Job Requirements

• 1. Bachelor's degree or above in Computer Science and Technology, Artificial Intelligence, Computer Science, Machine Learning, Big Data, Mathematics, or related fields.
• 2. In-depth understanding of large models and Agents; familiar with common LLM implementation paradigms, PE, RAG, and evaluation frameworks; familiar with mainstream Agent architectures (Multi-Agent, Context Engineering/Management, ReAct/PlanAct/CodeAct); knowledgeable about Agent-related interaction protocols (MCP, A2A, Function Call).
• 3. Familiar with methods such as Agentic RL; knowledgeable about relevant infrastructure (e.g., CodeRL, Reward System); capable of conducting SFT and RL training optimization for specific problems.
• 4. Strong academic vision and pioneering spirit; self-motivated to explore emerging technologies, with a proactive and feedback-driven mindset. A publication record at top conferences such as CVPR/ICCV/NeurIPS/ICLR/ACL/EMNLP/KDD is a plus.

Educational Background: Bachelor’s degree or above in Computer Science, Artificial Intelligence, Machine Learning, Big Data, Mathematics, or a related field.

Strong Understanding of LLMs and AI Agents: In-depth understanding of large language models and AI Agents. Familiarity with common LLM implementation patterns, prompt engineering (PE), retrieval-augmented generation (RAG), and evaluation frameworks. Knowledge of mainstream Agent architectures and approaches, including multi-agent systems, context engineering and management, ReAct, PlanAct, and CodeAct. Familiarity with Agent interaction protocols and interfaces, including MCP, A2A, and function calling.

Experience with Agentic Reinforcement Learning: Familiarity with Agentic RL and related infrastructure, such as CodeRL and reward systems. Ability to apply supervised fine-tuning (SFT) and reinforcement learning (RL) to optimize models for specific problems.

Research Acumen and Innovation: Strong research awareness, intellectual curiosity, and a willingness to explore emerging technologies. A proactive mindset, with the ability to think independently, take initiative, and provide constructive feedback. Publications at leading conferences such as CVPR, ICCV, NeurIPS, ICLR, ACL, EMNLP, or KDD are a strong plus.