Discussion - Topic: Giskard

This week's topic: Giskard

As described in Thoughtworks Technology Radar Vol. #28.

Giskard is an open-source tool designed to help organizations build more robust and ethical AI models by providing quality assurance capabilities with a focus on explainability and fairness. It facilitates cooperation between technical and nontechnical stakeholders, allowing them to evaluate models collaboratively and establish acceptance criteria based on bias avoidance and other essential quality metrics. Giskard ensures model outcomes are better aligned with business objectives and helps to solve quality issues before production deployment.

Zoom link will be added just before the event starts.

Discussion Resources :

Giskard’s open source framework evaluates AI models before they’re pushed into production by Romain Dillet
https://techcrunch.com/2023/11/14/giskards-open-source-framework-evaluates-ai-models-before-theyre-pushed-into-production/

Introducing the Giskard Bot: Enhancing LLM Testing & Debugging on Hugging Face by Hugging Face
https://huggingface.co/blog/JMJM/giskard-llm-testing-and-debugging-hf

Open-source library to test Large Language Models (LLMs)
https://www.youtube.com/watch?v=ATGa3x1ZNIE

Testing ML models Tutorial | Detect hidden vulnerabilities in your ML model
https://www.youtube.com/watch?app=desktop&v=8r9UfKN_NzA

Testing ML models Tutorial Part 2 | Generate a test suite for your ML model
https://www.youtube.com/watch?v=FAn2G1XoXX4

Testing ML models Tutorial Part 3 | Run test suites, apply slicing & transformation functions
https://youtu.be/aL3064qJo0w