Responsible AI Technical Guidance + Demos
[corporate talk alert!]
I was a PM on the responsible AI engineering team at Microsoft, I helped drive the strategy to develop and launch technical implementation guidance, policy, and playbooks for teams to deploy generative AI models and agentic systems responsibly. I also streamlined the process to educate and onboard customers onto our safety system to use with our Azure OpenAI Service. I've resigned from the team but still love talking about trust and safety when it comes to AI. Reach out if you want to chat!
Below are examples of external facing resources I've championed by working with research, policy, and design experts from across the company
Azure OpenAI Implementation
Organized and outlined technical recommendations and resources to help customers design, develop, deploy, and use AI systems that implement the Azure OpenAI models responsibly.
Overview of Responsible AI Practices for Azure OpenAI Models
Prompt Engineering Resources
Lead a multi-disciplinary team to develop a library of system prompt and prompt engineering resources based on common product use cases, domains, etc.
System message framework and templates
Azure OpenAI Access Framework
Updated the framework and process which is used to limit the access and use of Azure OpenAI (for appropriate use cases), including limiting access to the ability to modify content filters and/or abuse monitoring.
Limited Access to Azure OpenAI Service