Develop and construct platforms and tools tailored for Responsible AI principles such as fairness, security, and explainability across various model types (ML/DL/LLMs), data formats, and lifecycle stages.
Implement solutions and methodologies within existing AI projects to serve as safeguards against diverse model vulnerabilities, including toxicity and adversarial attacks.
Collaborate with business analysts, engineers, and stakeholders to ensure alignment of data science initiatives with Responsible AI principles.
Assist in devising and executing rigorous adversarial testing protocols for AI models to uphold Responsible AI standards.
Explore cutting-edge techniques, architectures, and methodologies to automate adherence to Responsible AI practices throughout the AI lifecycle, spanning from data preparation to model deployment and inferencing.
Establish monitoring systems to track model performance over time and institute mechanisms for regular model updates and maintenance.
Share insights and expertise across the organization to foster thought leadership and innovative strategies for addressing various aspects of Responsible AI. Technical and Professional Requirements:
Proficient in multiple machine learning (ML) and deep learning (DL) frameworks such as TensorFlow and PyTorch, along with expertise in programming languages like Python, R, SQL, Scala, and Julia.
Demonstrated proficiency in deploying Large Language Models (LLM) and Generative AI applications, with knowledge of techniques like Prompt Engineering Fine-Tuning (PEFT) and Retrieval-Augmented Generation (RAG). Experience in developing and deploying deep neural networks and ML models for intricate tasks. Familiarity with basic tooling for MLOps, LLMops, AI security, adversarial testing, and more. Preferred Skills: Technology->Artificial Intelligence->Artificial Intelligence - ALL Technology->Machine Learning->Accord Educational Requirements Bachelor of Engineering Service Line AI & Automation Service Line
* Location of posting is subject to business requirements