Researcher - Runtime Assurance

Conduct research and development on runtime assurance techniques for AI/LLM-enabled systems.

Design and implement a runtime assurance framework in Python to monitor, validate, and mitigate AI model uncertainties.

Develop and integrate uncertainty quantification, anomaly detection, and robustness evaluation techniques for LLMs and AI models.

Explore retrieval-augmented generation (RAG), AI observability frameworks, and runtime monitoring mechanisms for LLM-based decision-making with a focus on reducing hallucinations and improving factual consistency.

Conduct experimental evaluations to improve system robustness, performance, and adaptability.

Collaborate with AI/ML researchers and engineers to integrate runtime assurance techniques into AI development pipelines.

PhD in Computer Science, Artificial Intelligence, Machine Learning, or a related field.

Strong background in AI/ML, deep learning, LLM architectures, and uncertainty quantification.

Proficiency in Python programming, with experience in AI/ML frameworks (e.g., PyTorch, Tensor Flow, Hugging Face Transformers).

Familiarity with runtime monitoring, model evaluation, RAG, AI observability frameworks, and trustworthy AI research.

Experience with statistical analysis, formal methods, or anomaly detection techniques is an asset.

Strong research mindset with the ability to work independently and collaboratively in a fast-paced, research-driven environment.

Passion for AI system reliability, runtime assurance, and advancing trustworthy AI methodologies.