Integrating PAIG with AI Applications¶
This section is for the AI Application developers and integrators
You need to have some familarity of Large Language Models (LLMs) and programing languages like Python or REST.
PAIG easily integrates with any AI application (1), offering a variety of flexible options to ensure smooth connectivity and functionality:
- Read more about AI applications in the User Guide section.
-
Transparent Integration with LangChain
Tailored for AI applications developed with LangChain, this method affords effortless integration via the Privacera Shield library, ensuring automated policy enforcement. Given the expansive and feature-rich library of LangChain, which provides numerous features and components for AI application development—including support for Chains and Retrieval Augmented Generation (RAG) that have the potential to modify prompts—PAIG ensures that policies are consistently enforced, even on enriched prompts.
-
Python Library for Native AI Applications
This approach is optimized for AI applications natively developed using Python. Integrating the PAIG library directly with the AI application facilitates systematic policy enforcement. It also extends considerable flexibility to AI application developers, offering the latitude to invoke PAIG APIs in alignment with specific business and technical requirements.
PAIG Architecture¶
What Next?
-
Integrate With
-
Read More
-
How To