Intuit: Leveraging AI and ML

To help consumers avoid the tiresome chore of manually categorizing transactions. Intuit is employing machine learning to categorise transactions in QuickBooks. This calls for highly customised machine learning, similar to the Credit Karma service. This suggests to users how to raise their credit score depending on their inputs. QuickBooks and TurboTax are the most popular consumer finance products that are trademarks of Intuit. However, Intuit is forging ahead with aspirations to transform itself into an AI juggernaut. Thanks to recent advancements in machine learning and the purchases of Credit Karma and Mailchimp. Rajat Khare an employee at Intuit has been working throughout the QuickBooks ecosystem of products and services for the past 10 years. A video of Rajat Khare shows his talk on Running Intuit Quickbooks for 4M customers based on Graphql at GraphQL Asia.

AI-driven platform

Alon Amit, vice president of product management at the company, stated at the Data + AI Summit hosted by Databricks two weeks ago that “Intuit is evolving into an AI-driven expert platform.” “We’re here to help individuals succeed, and in the past, it meant assisting you with the formalities of filing your taxes and keeping your records. But now it has far more meaning.

Having a strong data architecture is necessary for creating these new data-driven solutions. The corporation preferred to have thousands of analysts, data scientists, and software engineers on the same page. With a single view of the data rather than creating multiple data systems for each of its projects.

When Amit and Manish Amde joined the business 3 years ago the architecture wasn’t even thought about. They are currently Intuit’s director of engineering. Both Amit and Amde worked at Origami Logic, which Intuit bought to aid in the development of their data and AI infrastructure.

The existence of numerous data silos was one of Intuit’s major problems. The hundreds of thousands of database tables containing decades’ worth of historical data were essential for helping Intuit’s analysts and data scientists understand client wants and develop new products. However, it was dispersed around the company which made it challenging to access. The simplest option is to copy the data, however, this has its own set of issues with accuracy and latency.


Intuit as the foundation for its new data architecture chose Databricks Delta Lake. It claimed to have discovered a happy medium between unmanageable data swamps and slow-to-adapt data warehouses by combining elements traditionally associated with a data warehouse (such as ACID transactions and quality guarantees) with the scalability and flexibility benefits of a data lake.

A tool called a data map is an important component of Intuit’s strategy. This data map tool is made up of three different types of data. These are the physical layer, the operational layer, and the business layer. The physical layer contains information about where data and the code that created it are located, the operational layer does the same for ownership and system dependencies, and the operational layer does the same for data classifications.

Intuit’s lakehouse architecture for real-time processing supports both Spark Streaming and Flink. The same data set is accessible to its analysts through Redshift and data science notebooks. The business is able to keep using Sagemaker as its main data science development tool, with MLFlow serving as a backup for automating machine learning tasks.

Also Read: Rajat Khare Founder of Boundary Holding – Artificial Intelligence company in Luxembourg

Related Articles

Leave a Reply

Your email address will not be published. Required fields are marked *

Back to top button
izmir escort
canlı casino siteleri casino siteleri 1xbet giriş casino sex hikayeleri oku