Tel Aviv – December 11, 2023 – Citrusˣ exited stealth today, announcing $4.5M in seed funding for their end-to-end AI validation and explainability platform.
The round was led by Canadian VC Awz with the participation of several angel investors.
As more industries adopt AI, the journey from model development to deployment in production is increasingly becoming a critical bottleneck for businesses. Machine learning (ML) models are often not understood outside of the data science team that developed them, posing challenges for validation, explainability, risk assessment, and legal approval. ML models are usually validated by generic open-source tools, but these are not well suited for increasingly complex and specialized use cases, or for explaining the models to non-experts. It typically takes 6-9 months for a model to make it to market.
Citrusˣ has built a platform for all stakeholders in an organization’s AI pipeline to ensure ML models are robust, explainable, and fair, enabling verification and risk mitigation by data scientists and risk officers, to ensure AI is being used responsibly and fairly according to regulatory requirements. Citrusˣ provides clear, accessible reports that explain what a model does, how it is performing, and ensure it can achieve business outcomes.
“In most organizations, the data scientists who built the models also test and validate them, and the rest of the business is blind to what’s happening with their machine learning,” said CEO Noa Srebnik. “By providing one unified platform for a whole company to use, Citrusˣ acts as the bridge that connects those mysterious ML models to the rest of the company through validation, monitoring, and reporting. Deployment and maintenance of ML models become smoother and faster, meeting regulatory demands, and helping to explain model decisions to end customers.”
The company’s patent-pending technology shows all elements and weak spots in a data space, identifies errors and vulnerabilities in an ML model, and highlights potential feature bias which leads to unexpected errors in the model.
“Collaborating across departments and managing multiple deployment models demands seamless communication, and that’s where Citrusˣ comes in”
Citrusˣ was founded by Noa Srebrnik (CEO) and Gilad Manor (CTO), after struggling to validate ML models using existing tools which slowed their models’ journey into production. Citrusˣ is already working with publicly listed companies in regulated industries such as finance and security.
“Machine Learning is expensive,” said Awz Founder and Managing Partner Yaron Ashkenazi. “Companies are not seeing returns on their investment when most ML models never make it into production, and even the good models can take close to a year to go live. By helping the whole organization to understand what an ML model is doing, and limiting risk, Citrusˣ is instilling the confidence required to go live, while ensuring optimal company performance.”
“Collaborating across departments and managing multiple deployment models demands seamless communication, and that’s where Citrusˣ comes in,” said a major Canadian bank working with Citrusˣ. “Its real-time report generation provides high-level insights about models which can be shared between model developers and oversight teams to provide a level of confidence and transparency in the model development process.”
Citrusˣ is part of the first cohort of the Awz X-Seed Hub, a pioneering, deep-tech, early-stage technology acceleration initiative. The Awz Active Capital team provides unparalleled funding; instrumental commercial connections; a two-year residency; and a comprehensive mentorship program.
To read the full article on SD Times, click here.
تعليقات