Daniel Nissani, Author

My background

I am a former mathematics teacher turned Responsible AI applied scientist and tester. In his current role, Daniel designs, implements, and executes ethical tests for preproduction models.

My expertise

I have expertise in fairness in machine learning and AI, natural language processing, and responsible AI.

When I'm not working

In my current role, I design, implement, and execute ethical tests for preproduction models.

Daniel's latest articles

An image of a woman and an AI agent standing together

Blog

Automating the Adversary: Designing a Scalable Framework for Red Teaming AI

Salesforce's trusted AI architecture for red teaming leverages automation to scale ethical AI testing, utilizing a tool called fuzzai to simulate diverse adversarial scenarios and enhance model robustness. By automating adversarial prompt generation and response validation, fuzzai helps secure AI interactions while reducing human exposure to harmful content.

4 authors

An illustration showing three people around a large computer monitor holding tools. Web analytics.

Blog

How Salesforce Builds Reproducible Red Teaming Infrastructure

Introduction Imagine you’re working on an AI product that can summarize customer success phone calls for training purposes. Your company’s product leverages large language models (LLMs) to summarize, synthesize, triage, and generate relevant…