Episode
Have a safe coffee chat with your documentation using Azure AI Services
with Maya Shavin
Building a custom documentation assistant using AI has become achievable with the help of GPT, LangChain and other AI tools. But how can we control the content quality of the coffee chat made to our document assistant, from the user to the assistant’s response? What options do we have to enhance the content safety in our question-and-answer flow, while scaling our project to handle further scenarios? Join my talk and let’s find out.
Chapters
- 00:08 - Welcome and Intro
- 01:20 - Document Q&A Chat intro
- 03:37 - Disclaimer - potentially sensitive content
- 05:00 - Content Moderation - Responsible AI
- 06:44 - Azure AI Content Safety
- 09:33 - Demo
- 16:43 - Integrate into an orchestration flow
- 17:49 - Azure OpenAI - Content Filters
- 19:14 - Azure AI Content Safety vs. Azure Open AI Content Filters
- 21:44 - Next Steps
- 23:29 - Wrap
Building a custom documentation assistant using AI has become achievable with the help of GPT, LangChain and other AI tools. But how can we control the content quality of the coffee chat made to our document assistant, from the user to the assistant’s response? What options do we have to enhance the content safety in our question-and-answer flow, while scaling our project to handle further scenarios? Join my talk and let’s find out.
Chapters
- 00:08 - Welcome and Intro
- 01:20 - Document Q&A Chat intro
- 03:37 - Disclaimer - potentially sensitive content
- 05:00 - Content Moderation - Responsible AI
- 06:44 - Azure AI Content Safety
- 09:33 - Demo
- 16:43 - Integrate into an orchestration flow
- 17:49 - Azure OpenAI - Content Filters
- 19:14 - Azure AI Content Safety vs. Azure Open AI Content Filters
- 21:44 - Next Steps
- 23:29 - Wrap
Have feedback? Submit an issue here.