Aporia exits Stealth with $ 5 million for surveillance platform to ensure AI integrity


TEL AVIV, Israel, April 6, 2021 / PRNewswire / – Aporia, the production observability SaaS platform for machine learning, left the stealth today and announced the launch of the first customizable monitoring platform for machine learning models with support for full load of private and public clouds. The company also revealed $ 5 million in start-up funding for Vertex Ventures and TLV Partners and is already used by multi-billion dollar companies. Aporia enables data scientists to quickly and easily create their own monitors, so they can track the performance of their machine learning models, ensure data integrity, and deliver responsible AI.

Companies around the world are investing up $ 50 billion annually on AI adoption, but a lack of observability and a poor ability to quickly spot problems in ML models as they run in production are undermining their investment.

The only way to know if a complex system is working as expected is to monitor it. Nonetheless, machine learning models are difficult to follow both technically and conceptually, as they rely on real world data to make accurate predictions about the future. Machine learning models can work fine in the experimentation phase, but they start to drift over time due to technical changes to databases and APIs – or more often due to ever-changing data and new events happening in the real world. Something as routine as a business expanding into a new market, or as dramatic as the Covid-19 pandemic can have significant implications for a model’s performance.

When machine learning models fail, customers and businesses suffer the consequences. Predictions based on faulty data are often wrong, leading to unexpected results such as a customer being presented with bad recommendations or accidental denial of a loan application. This, in turn, can result in lost revenue or expose companies to complaints of discrimination and injustice. Without proper monitoring, it can be months before a company even notices that its models have stopped making accurate predictions.

“AI needs guardrails,” says Liran hason, CEO of Aporia. “Businesses need to be confident in their machine learning models, and the only way to do that is through robust monitoring to make sure they’re doing what they’re supposed to do. “

With Aporia, data scientists can create custom monitors for their machine learning models with just a few clicks and set alerts of different severity to send via email or to sources like Slack. Aporia’s monitors are extremely flexible, allowing data science teams to monitor the right things for their own models and business cases.

Aporia can be installed with a few lines of code and monitors asynchronously, handling billions of daily prediction workloads without impacting latency. The user interface is comprehensive, clear and simple, making it easy to create, maintain and modify monitors. Once Aporia’s platform reveals a problem, data scientists can often quickly investigate the cause of the problem and decide how to fix it, whether through a change in logic, a bug fix, or a recycling of the model. ML if necessary.

Concerns about data security and regulations make many companies reluctant to adopt public cloud monitoring tools. Along with its deployment in the public cloud, Aporia offers an innovative “managed on premise” solution, offering peace of mind to large enterprises and enterprises with high demands on privacy and data security.

Liran hason, the founding CEO of Aporia, is a veteran of the IDF’s elite 81 intelligence unit. He was an early employee of Adallom (acquired by Microsoft), where he led the ML production architecture, serving millions of users. Prior to launching Aporia, Hason was part of the Vertex Ventures investment team and participated in over 30 investments including Axonius, Spot.io and others.

“Companies are struggling to monitor their AI in ways that matter to their machine learning model and use case,” Hason said. “Aporia makes monitoring simple, fast and secure, incorporating engineering and DevOps best practices into the new realm of MLOps and ensuring that data science teams can maintain their performing models with accuracy and fairness. “

Emanuel Timor, Managing Partner of Vertex Ventures, says “AI adoption is skyrocketing and requires the right technology stack to meet the new challenges that come with it. Aporia is an essential part of the new MLOps stack, filling a critical gap in AI production readiness. . “

Rona segev, Founding Partner at TLV Partners: “Tracking production workloads is a well-established software engineering practice, and it is high time machine learning was monitored at the same level. The Aporia team has solid experience in production engineering, which sets its solution apart. so simple, safe and robust. “

Media contact
Lazer cohen
WestRay Communications
[email protected]
+1 347-753-8256


Related links


Leave A Reply

Your email address will not be published.