Startup spearheading the next phase of AI

In today’s world, it’s no secret that data is central to running an effective business. The collection and analysis of data is only growing in importance as brands continue to leverage it to improve business operations and connect with customers. A new startup, Loopr, aims to spearhead the next wave of data-driven technology that will power the Artificial Intelligence models of the future.

The brainchild of founder and CEO, Priyansha Bagaria, Loopr is a cloud-based data labeling platform that uses semi-automated tools and a dedicated team of annotators to collect and organize different raw data sets (text, image, or video) that are then used to power AI models in a variety of industries.

“The potential for AI models is endless,” said Bagaria. “Once an AI model has been trained with a dataset labeled with metadata information, they can be used to create innovative solutions by businesses to solve key problems and increase revenue, optimize costs and much more.”

Loopr plans to recruit an additional 10 to 15 employees to support the startup’s continued growth in 2022. In the future, Loopr plans to expand into different industries and continue to develop unique AI models for diverse applications.

“We are a female-led company and our team of data annotators comprises only women,” said Bagaria. “We embarked on this journey to both empower women and create a company that develops real, impactful solutions for organizations. We’re excited to explore new industries and continue to use data and AI to improve company processes and elevate customer and consumer outcomes.”

Once data has been collected and organized into the Loopr platform, the process of training AI models occurs in several ways.

Take the manufacturing industry, for example, where Loopr uses AI to detect defects in products. The AI model is trained to detect imperfections by inputting thousands of diverse images of flawed products, where the defect in each image is marked and labeled. Then, the model learns to identify and classify any type of defect by itself in any scenario.

“Look at wind turbines — a product that must be absolutely perfect to function properly,” said Bagaria. “With massive, high-end turbines made of steel, it can be challenging to identify defects with the naked eye. Loopr’s AI models analyze thousands of images of defects in turbines, then use this information to identify imperfections that humans could not. It is tremendously cost and time-saving to have an AI model find issues down to a pixel granularity.”

Loopr also has pre-built AI models that businesses can deploy so organizations don’t have to build models from scratch.

Looking ahead, Loopr plans to recruit an additional 10-15 employees to support the startup’s continued growth in 2022. In the future, Loopr plans to expand into different industries and continue to develop unique AI models for diverse applications.

This is part of a regular guest column written by the Technology Association of Oregon in the Portland Business Journal.