Zohar Bronfman, CEO and Co-founder of Pecan, had no inkling of a career in tech during his years of military service in the Israeli army, or while earning two PhDs in computational neuroscience and another in the history of philosophy. His journey from soldier to scholar to CEO of a rapidly growing AI/ML tech startup is as unusual as it is inspiring.
Pecan is an AI-powered analytics platform that leverages AI and ML to provide advanced analytics and predictive modeling solutions. It utilizes AI techniques to automate and enhance various aspects of the analytics process, enabling businesses to gain valuable insights and make data-driven decisions without a need for existing data science proficiency.
I sat down with Zohar to talk about his transition to startup co-founder, what differentiates Pecan from other analytics providers, and his company’s relationship with Microsoft solutions. The first question, however, was how to pronounce his company’s name: “PE-can,” or “peh-CAHN”?
“We have a company motto,” Zohar laughs. “It doesn’t matter how you pronounce it, as long as you buy it.”
Rapid growth brought on by recognizing a market need
Zohar is quick to admit that his path to tech startup founder and CEO is atypical. During his three years in the Israeli army, he was stationed in the country’s largest intelligence unit but assigned to the non-technological division. His subsequent academic pursuits were also far from his eventual career in AI/ML, but they set the stage for what was to come.
“I met Noam, our cofounder and CTO, on the very first day of our Masters in Neuroscience in Tel Aviv University,” Zohar recalls. “We became interested in the field of machine learning and data science and how models can emulate brain processes and predict human behavior.”
Zohar and Noam would then launch Pecan and develop its automated machine learning (autoML) platform, designed to be accessible to businesses without requiring expertise in data science or programming. Capable of automatically generating the best possible model for a given problem based on the data provided, the solution is particularly well-suited for businesses that don’t have dedicated data science teams or that need to move quickly to implement machine learning models.
“Implementing these models is a huge problem,” Zohar says. “A ton of money is thrown down the drain because of a lack of capacity to find and train machine learning talent. Our rapid growth speaks to the need in the market and the value potential customers see in our platform.
“There are barriers to entering our domain in terms of competition, especially concerning data automation. Our main advantage is that our customers do not require expertise in data science, which is enabling them to use our services without any prior knowledge in the field.”
“Our mission is to put the power of data science in the hands of data and BI analysts,” he continues, “and we’ve recently opened our platform for anyone to sign up and try their hand at automated predictive analytics. We’re excited to see more and more data analysts take on predictive modeling and help drive business success.”
Innovating in the field of AI and ML
Pecan’s use of several AI techniques to automate and enhance their analytics process—including data prep and feature engineering, templates for models that address specific business problems, and SQL-driven modeling with autoML—enables their customers to gain valuable insights and make data-driven decisions. Zohar says one of the biggest problems Pecan addresses, other than reducing the need for trained data scientists to implement models, is properly gathering data with which to build models.
“A company may not have data properly prepared, structured, or engineered for machine learning,” Zohar explains. “We built automation around taking raw data and transforming, collating, restructuring, cleansing, and engineering it so it can be meaningfully fed into the ML algorithms.”
Zohar says this patented technology opens a new field of innovation with the potential to fill a wide array of applications, including customer segmentation, churn prediction, and personalized marketing. One of Pecan’s goals is to develop models that are not only accurate but also transparent and explainable. Zohar believes this will help differentiate Pecan in the market as more businesses adopt AI and ML.
“There are barriers to entering our domain in terms of competition,” Zohar says, “especially concerning data automation. Our main advantage is that our customers do not require expertise in data science, which is enabling them to use our services without any prior knowledge in the field.”
A close partnership with Microsoft forges a path to mentorship
Zohar says another key to Pecan’s market penetration is their close partnership with Microsoft. Through their connections with other Microsoft clients, Pecan can add the value of their platform to existing Microsoft software, helping them to gradually expand their customer base.
“We are extremely proud about working with Microsoft,” Zohar says. “We get amazing support on both the technological and go-to-market sides.”
Pecan’s platform is built on Azure Databricks, adapts an Azure infrastructure and utilizes Apache Spark in in Azure HDInsight for automation. The company trains and deploys its ML models on Azure Machine Learning, integrates with Azure Data Factory, and houses its data on Azure Data Lake, allowing them to store large amounts of data at a low cost. In addition, Pecan also integrates with Power BI, allowing their users to visualize and analyze their data.
“Thirty percent of data analysts use Power BI,” Zohar explains. “We already see our customers integrating predictions from Pecan into their Power BI dashboards and tools. BI now includes more than analyzing what happened in the past—it’s also analyzing and communicating what’s going to happen in the future. Across the board, these are exciting opportunities to expand the meaning and impact of business analytics.”
Zohar says Pecan’s membership in the Microsoft for Startups Founders Hub has also paid dividends for the company.
“We’ve been working closely with the Founders Hub for years, and they have always been there for us with content, mentorship, and programs that have helped us grow individually and make connections,” Zohar explains. “It’s reached the point where now I’m being asked to provide mentorship, so I’m really proud of the collaboration we’ve generated.”
Microsoft for Startups Founders Hub members receive Azure cloud credits that can be used toward Azure OpenAI Service or OpenAI to help build their product. Sign up now to become a member.