AI Researcher (Alignment & Agentic Safety)

About Will

Wizard Intelligence Learning Lab (Will) is an entrepreneurial laboratory incubated by WizardQuant. Will is dedicated to building ASI for Sci-Tech. Bringing together exceptional AI talent, Will focuses on developing general-purpose AI systems for science and technology, with foundation models at the core of its work. Through an industrial-grade R&D approach, we advance model capabilities and explore their applications in depth, while pushing the frontier of AI across models, systems, and real-world use. What we seek to build is not a replacement for people, but a foundational capability platform that supports scientific discovery, technological innovation, and knowledge creation.

Job descriptions

As a founding member of the Alignment team, you will help ensure that increasingly capable AI systems remain aligned with human intent as they are deployed in complex, long-horizon, real-world settings. You will research and develop scalable alignment methods that continue to work as model capabilities advance, ensuring that agent behavior remains consistent with human goals, legal and regulatory requirements, and broadly accepted societal norms.

Responsibilities

• Design and execute alignment-focused experiments across subjective, contextual, and long-context settings.
• Develop evaluations that measure instruction-following, alignment, and risk in agentic systems.
• Build tools and methodologies for studying robustness, faithfulness, truthfulness, and reliability.
• Research scalable supervision mechanisms and new human-AI interaction paradigms.
• Improve calibration between model capabilities, uncertainty, and risk.
• Curate, evaluate, and improve alignment-related datasets.

Qualifications

• Strong research background in computer science, machine learning, cognitive science, data science, or related disciplines.
• Excellent engineering skills, particularly in large-scale machine learning systems.
• Deep understanding of alignment, evaluation, and safety concepts.
• Proficiency in TypeScript, Python and modern development tools.
• Experience building research tooling, data collection systems, or visualization platforms.
• Strong communication and collaboration skills.
• Passion for trustworthy AI in scientific and high-stakes environments.