19: Azure Cognitive Services: The Cool and the Creepy

Development Best Practices

After a wonderful summer break, it's time for Episode Nineteen of the Development Best Practices podcast, brought to you by ILM Professional Services. In this episode we continue our talk with Sam Nasr, Microsoft MVP and Organizer of the Cleveland .NET User Group, about Azure Cognitive Services. Sam talks through the cool and the creepy in cognitive services in general, as well as pricing and how you can get feedback to Azure on errors or improvements to be made.
1:15 How creepy is face detection?
2:40 Best practices in data set size
4:15 Our biggest "wow" moments in cognitive services
6:45 Success stories
9:20 Azure Cognitive Services pricing model
10:45 Getting feedback to Microsoft
11:45 GIGO and responsible usage
Azure Cognitive Services: https://azure.microsoft.com/en-us/services/cognitive-services/#api
Responsible AI investments and safeguards for facial recognition: https://azure.microsoft.com/en-us/blog/responsible-ai-investments-and-safeguards-for-facial-recognition/
Azure Cognitive Services support and help options: https://docs.microsoft.com/en-us/azure/cognitive-services/cognitive-services-support-options
All Sam's contact Links: https://linktr.ee/samnasr
The Development Best Practices Podcast is brought to you by ILM Professional Services. Sam talked about how, like with so much in the development world, the response to many Azure Cognitive Services questions is "It Depends". This is where the truly great consultants excel. If a consultant gives you a cookie cutter answer, you should expected a lot of crummy results. A good consultant will say "it depends" and then LISTENS, creating a custom solution for you. That's what we do at ILM - you'll never get a cookie-cutter solution. Please visit ilmservice.com today to find out why, and how to work with some of our consultants!
Oh, and as we're about to leave our teens in the episode count, I am so grateful for the feedback and topic suggestions we've received. If YOU have any feedback on this episode, want to suggest a topic, or need anything else, please email us at [email protected].
Thanks for listening, and have a great day!