AI Safety Researcher

Apply Location

Beijing/Singapore

Application Deadline

8 November 2026, 23:59 GMT+8

Application Process Our hiring process has 4 phases: - Written application form - Online written test - 2 online interviews - 2-day in-person work trial *An additional coding test will be included as part of the online written test stage.

Background

Concordia AI is committed to identifying and addressing the most pressing safety risks arising from frontier AI development. As large model capabilities advance rapidly and agent frameworks continue to evolve, risks such as AI misuse and loss of control have become urgent scientific challenges that demand solutions. We are looking for talented individuals with rigorous scientific research skills to join us and engage deeply in building safety benchmarks, advancing red-teaming attack and defense algorithms, and developing quantitative risk modeling. Together, we will produce top-tier research outcomes and drive the development of AI technology along a safe and controllable path.

Responsibilities

As a core team member at Concordia AI, you will drive frontier safety research for large models and AI agents. Your work will encompass building robust AI safety benchmarks, researching red-teaming attack and defense algorithms, developing risk modeling, and devising mitigation strategies. You will also engage in translating cutting-edge research into real-world applications to strengthen the safety foundation of AI systems.

• AI Safety Benchmark Development: Independently design and build capability and safety benchmarks for large models, AI agents, and multi-agent systems. Curate high-quality evaluation datasets, develop automated testing pipelines, conduct systematic evaluations across diverse models and agents, and produce technical reports or academic papers.
• AI Red-teaming Attack and Defense Algorithm Research: Track state-of-the-art adversarial techniques in AI safety. Conduct red-teaming research on large models, AI agents, and multi-agent systems to identify effective attack and defense algorithms, discover critical vulnerabilities and risks, and author technical reports or academic publications.
• AI Risk Modeling: Build quantitative risk modeling for frontier risk areas, including cybersecurity, biological, chemical, harmful manipulation, and loss-of-control risks. Conduct comprehensive risk assessments using evaluation data, expert forecasts, and real-world incident case studies, and produce assessment reports or academic papers.
• AI Risk Mitigation Research: Investigate advanced methods in AI alignment, monitoring, interpretability, and agent/multi-agent safety. Explore, prototype, and validate innovative safety solutions to mitigate risk, and produce technical reports or academic papers.
• Research Translation & Impact: Integrate research outcomes into Concordia AI’s Frontier AI Risk Monitoring Platform and commercial safety evaluation and remediation services. Leverage empirical data to support safety governance reports, standards recommendations, and policy briefs.

Note: You are not expected to work across all of these research areas simultaneously. We welcome candidates who specialize in one or two of these tracks.

Qualifications

Required

• Alignment with Concordia AI’s mission, along with a strong sense of responsibility, risk awareness, and a proactive mindset; Ability to work independently while collaborating effectively within a team.
• Bachelor’s degree or above.
• Exceptional scientific reasoning skills, ability to break down complex research problems, and strong learning capacity to continuously track and apply the latest advances in the field.
• Strong foundation in data structures and algorithms, with proficiency in Python for algorithm design and data analysis.
• Solid understanding of large model and AI agent architectures; Practical experience with large model deployment, API usage, and harness development; Hands-on experience in large model/agent safety evaluation and risk analysis.
• Excellent logical reasoning and technical writing skills to draft research proposals and experimental reports.
• Strong Chinese reading and writing skills, with proficient spoken Chinese for smooth communication.

Preferred

• Degree in Computer Science, Artificial Intelligence, Machine Learning, or a related field.
• 2+ years of research or industry experience focused on large models, AI agents, or AI safety.
• Top-tier conference publications in large models, AI agents, or AI safety.

Location

Beijing/Singapore

For detailed information, please see: https://mp.weixin.qq.com/s/QIunhlvvq3OKpIDZE-1ZiA

Apply