Research Engineer II (Computer Science / Cyber Security / Artificial Intelligence)
Core
Researching environment-aware, task-oriented access control and safety mechanisms for autonomous LLM agents to ensure secure and reliable AI ecosystems.
Role type
Research Engineer II (AI Safety & Access Control)
Builds
Secure, resilient AI agent ecosystems and research prototypes for trustworthy machine learning.
Domain
Artificial Intelligence, Cyber Security, Green Digital Technologies
Deliverable
production ML models | research
Required skills
LLM agent safety, access control design, least-privilege policy generation, runtime adjudication, reinforcement learning, fine-tuning large language models, Python, C/C++, Java, academic writing
Preferred skills
RL experience, tool use in multi-step reasoning, parameter-efficient fine-tuning methods (LoRA)
Technologies
PyTorch, OpenClaw, WildClawBench
Responsibilities
Design and implement a Profile-and-Policy Agent for structured authorization context; Develop and evaluate a Runtime Guard Agent for adaptive controls; Construct attack payloads and benchmarks; Implement optimization methods like reinforcement learning; Contribute to manuscript drafts for top-tier venues.
Seniority
Mid-level Research Engineer
