Site icon AIT365

Kolena Launches AI / ML Model Quality Platform to Enable Rapid, Accurate Testing and Validation of AI Systems

Kolena

Innovative Startup Building Toward End-to-End Quality Assurance, Testing and Monitoring Solution to Help Organizations Create Trustworthy Systems Across All AI / ML Modalities

Kolena, a pioneering leader in AI and machine learning model testing, announced the wide launch of its cutting-edge AI / ML Model Quality Platform following a 24-month closed beta with a select group of early access customers that included Fortune 500 companies, leading AI startups, government organizations and European AI standardization institutes. Kolena expanded its platform’s capabilities over the course of the beta to include comprehensive pre-training data quality features in addition to its powerful model testing and validation functions, and is working toward an end-to-end AI quality solution that incorporates data quality, model testing and A/B testing, along with monitoring for data drift and model degradation over time.

Kolena customers conducted tens of thousands of tests on AI / ML models during the beta, allowing Kolena to refine and expand its platform’s capabilities while saving customers thousands of engineering hours and radically increasing the accuracy of their AI systems.

The wide launch of Kolena’s platform comes at a critical time for the AI industry, as problems such as historically inaccurate imagery created by generative AI models, misidentification of pedestrians by self-driving automobile models, hallucinations and other failures continue to weigh on adoption trends for the technology. The launch will enable Kolena to expand its leadership position in establishing best practices and standards for AI model quality, and to broaden its focus on outreach, education and training for the global AI / ML community.

Kolena Co-Founder and CEO Mohamed Elgendy said, “The Kolena AI / ML Model Quality Platform represents a cutting-edge approach to unlocking AI adoption for both enterprises and consumers by empowering AI builders to launch their models with confidence, knowing that each one will perform flawlessly in real-world scenarios. We are emerging from beta with a platform that addresses the full spectrum of testing and quality assurance needs, across any industry, domain, size or ML modality, and now incorporates pre-training data quality functions.”

Also Read: OutSystems Announces Its Latest AI Innovation: AI Agent Builder

Mr. Elgendy continued, “Our unique approach allows for detailed testing, revealing hidden flaws and strengths and enabling data scientists at startups and Fortune 500 companies alike to focus on what matters most. As standards bodies and governments around the world place more emphasis on AI model testing and monitoring, Kolena is leading the way to transform ML development into a true engineering discipline and take the guesswork out of AI deployment. Our platform provides an invaluable tool to help organizations’ technology meet the highest standards of quality and reliability, in order to translate their AI systems’ potential into reality.”

Olgert Denas, Senior Manager of Machine Learning at Kolena customer John Deere, said, “Adopting Kolena’s cutting-edge AI / ML model testing framework has significantly elevated our approach to AI development. Kolena’s transformative model quality framework has empowered us to standardize the quality of our models across the board, streamlining our development cycle and enabling us to deliver superior models much faster than ever before. With Kolena, we’ve accelerated our workflow and overall model development cycle and ensured that all team members and stakeholders have instant access to vital information. Kolena has become an essential tool in our arsenal, transforming our operations and allowing us to innovate with assurance and speed.”

Kolena’s AI / ML Model Quality Platform

Kolena’s AI testing and validation solution helps developers build safe, reliable, and fair AI systems by allowing them to instantly develop razor-sharp test cases from the data sets powering their models. This enables them to rapidly scrutinize AI / ML models in the precise scenarios those models will be entrusted to handle in the real world, rather than having to rely on aggregate statistical metrics that can obscure a model’s performance on high-value tasks.

By integrating Kolena into the AI development lifecycle, organizations can benefit from streamlined model testing processes, enhanced model reliability, and improved accountability in AI applications. The company’s solutions support a proactive approach to model management, enabling developers and data scientists to address potential problems before they impact decision-making processes.

Additionally, Kolena’s automated testing workflows reduce model validation efforts from weeks to minutes by eliminating the need for time-consuming experimentation-based testing, helping teams launch their models with far greater speed and efficiency. This approach also provides insights into model performance and helps identify issues related to data drift, model degradation, bias and other problems over time. The company’s testing capabilities span all AI / ML modalities, from computer vision, language and structured data to generative AI, LLMs and multi-modal models.

Kolena‘s powerful testing and validation solution is at the core of the holistic AI quality platform the company is building to tackle performance issues across the AI / ML development lifecycle. The platform, which currently includes data quality and pre-deployment model testing, will soon include post-deployment model monitoring, forming a comprehensive platform that ensures high quality models are delivered to end users while eliminating failure modes like biases, safety / security failures, hallucinations, and others.

SOURCE: PRNewswire

Exit mobile version