DeepLearning.AI has released a new, free short course titled “Safe and Reliable AI via Guardrails”. The course is introduced in partnership with Guardrails AI and is taught by its founder Shreya Rajpal.
The course will help the learner explore, and mitigate some of the common failure modes in LLM-based apps, such as hallucinations and exposure to sensitive information, among other undesirable and harmful outputs. The course will also help the learner implement ‘guardrails’ to prevent these failure modes, by accessing ‘pre-built’ guardrails on GuardrailsAI.
“This course will show you how to build robust guardrails from scratch that mitigate common failure modes of LLM-powered applications like hallucinations or revealing personally identifiable information (PII),” said DeepLearning.AI.
To help the learner understand these concepts through a real-world example, these guardrails will be implemented on a small pizzeria’s ‘RAG-powered customer service chatbot’. The course will help set up a guardrail to limit the chatbot’s output relevant to the pizzeria and use a ‘Personal Identifiable Information’ guardrail to detect sensitive information in both input prompts and output responses.
The course only requires basic Python knowledge as a prerequisite. In a preview video for the course, Andrew Ng said, “I think they’ll (guardrails) become increasingly important in the AI applications stack, especially in industries with stricter safety, privacy, or regulatory requirements.”
Last week, DeepLearning.AI released a short course titled “LLMs as Operating Systems: Agent Memory” to help learners build agents. The course helps learners build an agent that can successfully manage the memory and context of an LLM using Letta’s open-source framework.
Currently, DeepLearning.AI provides over 50+ short, and free courses focusing on some of the best techniques for the safe, and efficient deployment of generative AI products.
DeepLearning.AI offers dedicated courses and specialisations on broader topics like data engineering, machine learning, deep learning, and so on. Andrew Ng’s efforts in democratising AI education have reached over 7 million students, who are using the platform to future-proof their skills. Interestingly, he’s also the co-founder of Coursera and was recently inducted into Amazon’s board of directors.
The post Andrew Ng Releases New Short Course on Building Safe & Reliable AI appeared first on Analytics India Magazine.
