Become a sponsor to Giskard
We are building the first holistic Testing & Evaluation platform for AI agents. We help AI agent developers increase the efficiency of their AI development workflow, eliminate risks of AI agent failures and ensure reliable & secure AI agents.
We are a team of engineers & researchers on AI Quality & Security who have been working on this topic since 2021. While we are embracing about the new Generative AI revolution, we acknowledge the risks involved and the need for testing to control risks of prompt injections, hallucinations, etc.
We believe crucial to have independent third-party evaluations to control the risks of AI agents. These evaluations, conducted by separate entities from the AI developers, provide important checks and balances to ensure responsible controls of the AI ecosystem.
By sponsoring our open-source project, you can help bring agentic AI into the age of Trust!
Support the development of Open-Source Testing & Evaluation methods for AI systems
Meet the team
-
Jean-Marie John-Mathews jmsquareCo-founder & co-CEO of Giskard | Ph.D. in AI Ethics, Ex-Thales data scientist
-
Alex Combessie alexcombessieCo-founder & co-CEO of Giskard | Ex-Dataiku AI engineer & data scientist
-
Matteo mattbitCTO @ Giskard
-
Inoki InokinokiSoftware Engineer @ Giskard
-
Blanca Rivera Campos BlancaRiveraCamposCommunity & Growth Manager @ Giskard
-
Pierre Le Jeune pierljML Research @ Giskard
-
Kevin Messiaen kevinmessiaenSoftware Engineer @ Giskard
-
-
Henrique Chaves henchavesDeveloping data products 😊
Featured work
-
Giskard-AI/giskard-oss
🐢 Open-Source Evaluation & Testing library for LLM Agents
Python 5,327 -
Giskard-AI/awesome-ai-safety
📚 A curated list of papers & technical articles on AI Quality & Safety
-
Giskard-AI/phare
Phare is a LLM benchmark that evaluates models across key AI security & safety dimensions
Python 13 -
Giskard-AI/giskard-client
API Client to interact with the Giskard platform using code 💻
Python 7 -
Giskard-AI/giskard-agents
A lightweight library that orchestrates LLM completions and agents in parallel workflows
Python 5 -
Giskard-AI/flare
Phare Benchmark Runner
Python 4