Research Lead – Hybrid AI Security

Lenovo Digital Trust Lab is looking for a Research Lead to join our Hybrid AI Security Team—a visionary thinker and technical expert to drive innovation at the intersection of LLM security, agentic systems, and multi-agent orchestration. This role is central to shaping the future of secure, scalable, and trusted AI systems.

The ideal candidate brings deep expertise across the AI stack—from foundational layers (hardware, compute, model architecture) to orchestration frameworks, multi-agent communication protocols, and task-level coordination—while maintaining a strong understanding of emerging threats and defenses in this rapidly evolving space.

Job Responsibilities:

• Serve as the technical and research leader for the Hybrid AI Security domain.
• Lead cutting-edge research initiatives in areas such as:
• LLM security and model misuse
• Agentic systems and agent security
• Multi-agent orchestration and trust frameworks
• Prompt injection, sandboxing, data leakage, and control mechanisms
• Act as a hands-on contributor: from ideation and prototyping to publishing and patenting.
• Stay ahead of the curve by tracking emerging threats and novel attack surfaces in AI-native environments.
• Drive cross-functional collaboration with engineering, product, and security teams to integrate research outcomes into real-world solutions.
• Build and nurture strategic relationships with academia, industry leaders, and standardization bodies.
• Represent the lab at global conferences, contribute to open-source and academic publications, and serve as a thought leader in the space.

Minimum Requirements:

• MSc or PhD in Computer Science, AI/ML, Cybersecurity, or a closely related field.
• 8+ years of experience in cybersecurity (offensive, defensive, or applied research roles).
• 5+ years of experience in AI, including deep learning, language models, or agent-based systems.
• Proven experience in LLM security, agentic systems, or multi-agent security—through research, publications, patents, or applied systems.
• Deep familiarity with the full LLM stack: from infrastructure and model architecture to inference, prompt engineering, fine-tuning, agent frameworks, and API-based deployments.
• Demonstrated ability to translate complex research into practical, scalable, and secure solutions.
• Strong publication record or recognized industry contributions in relevant domains.

Preferred Requirements:

• Experience with adversarial ML, AI red teaming, or secure-by-design AI systems.
• Familiarity with frameworks like LangChain, AutoGen, OpenAI tools, or equivalent agent orchestration systems.
• Deep understanding of AI-related threat modeling, attack vectors, and defensive architecture.
• Experience in cross-disciplinary teams or research collaborations.
• Background in standardization, policy engagement, or ethics in AI and security.