.NET AI Community Standup: What’s New in .NET AI-Samples


Summary

The video showcases a team call where participants introduce themselves and discuss diverse topics related to AI samples, tutorials, and model evaluation. They delve into updating features, running user tests, and exploring metrics in Valproject. The session also covers responsible AI practices, testing scenarios, and potential applications, while emphasizing feedback and documentation for continuous improvement.


Introduction and Greetings

Participants exchange greetings and introductions at the beginning of the call.

Introducing Team Members

Team members introduce themselves, their roles, and briefly discuss the topics to be covered in the call.

Topics to be Discussed

An overview of the topics to be covered, including work updates and a recap of a recent event.

Recap of Recent Event

Discussion about a recent event, sharing highlights and encouraging viewers to explore further.

AI Samples and Tutorials

Details about AI samples and tutorials, including quick start tutorials and the use of semantic carol and Ashop ASK.

Adding Content to AI Samples

Discussion about adding more content and features to the AI samples, including tokens, embeddings, and local models.

Building Tools for Model Evaluation

Demonstration and discussion about tools for model evaluation, including code snippets and evaluation metrics.

Evaluating Different Models

Explanation of the process of evaluating different models, including ground truth evaluation and model comparisons.

Future Plans and Community Feedback

Discussion about future plans, community feedback, and potential additions to the AI samples repository.

Analyzing User Tests in Valproject

Running and analyzing user tests in Valproject, including performing various actions and showing metrics related to telemetry and user stories.

Updating Tables and Output of Models

Discussing the process of updating tables and outputting models in real-time in Valproject, focusing on user story analysis.

Exporting Metrics and Integration

Exploring options to export metrics, integrate with openTelemetry, and showcase data on dashboards in Valproject.

User Story Capabilities and Generation

Discussing user story capabilities, generation, coherence, relevance, and analyzing input data in Valproject.

Prompt Evaluation and Tailored Scenarios

Examining prompts for evaluation, tailored scenarios for user stories, and the importance of specific examples in Valproject.

Model Groundedness and Data Relevance

Analyzing model groundedness, data relevance, and the impact of user story data on models in Valproject.

AI Evaluation and Quality Metrics

Exploring AI evaluation, quality metrics, metrics dashboard, and custom evaluators in Valproject.

Responsible AI and Harmful Content

Discussing responsible AI practices, testing for harmful content, and examples of misleading AI applications in Valproject.

Testing with Different Configurations

Testing AI models with different configurations, evaluating coherence, relevance, and performance in Valproject.

AI Testing Scenarios

Exploring AI testing scenarios, evaluating models, and potential applications for specific AI models in Valproject.

Speech Command Integration

Discussing integrating speech commands, adapting models for evaluation, and exploring potential applications in Valproject.

Feedback and Documentation

Encouraging feedback, discussing model testing and modifications, and the importance of documentation in Valproject.

Command Line Applications

Exploring command line applications, utilizing Spectre in .NET, and showcasing unique functionalities for applications in Valproject.

Conclusion and Future Plans

Wrapping up the session, discussing future topics, and hinting at upcoming presentations in Valproject.

Logo

Get your own AI Agent Today

Thousands of businesses worldwide are using Chaindesk Generative AI platform.
Don't get left behind - start building your own custom AI chatbot now!