In a fast-paced environment, how does our small data science team continually solve our company’s and our customers’ biggest challenges?
At Razorpay, our mission is to be a one-stop fintech solution for all business needs. We power online payments and provide other financial solutions for millions of businesses across India and Southeast Asia.
Since I joined in 2021, we have acquired six companies and are expanding our product offerings.
Although we are growing rapidly, Razorpay competes with much larger organizations with significantly more resources to build data science teams from scratch. We needed an approach that leveraged the expertise of our 1,000+ engineers to build the models they need to make faster and better decisions. Our vision for AI was fundamentally based on empowering our entire organization with AI.
Driving rapid machine learning and AI experimentation in financial services
Given our goal of putting AI in the hands of engineers, ease of use was at the top of our wish list when evaluating AI solutions. They needed the ability to climb quickly and explore without too much tedious holding.
Regardless of someone’s background, we want them to be able to get answers quickly out of the box.
AI experimentation like this took a whole week. We have now reduced this time by 90%, meaning we get results in just a few hours. If someone wants to jump in and get an AI idea moving, it’s possible. Imagine that time savings multiplied across our entire engineering team – that’s a huge boost to our productivity.
This speed has allowed us to solve one of the most difficult business challenges for customers: fraudulent orders. In data science, timelines are usually measured in weeks and months, but we achieved this in 12 hours. The next day we went live and blocked all malicious orders without affecting a single real order. It’s pretty magical when your ideas come to life so quickly and have a positive impact on your customers.
“Playing” with the Data
When team members load data into DataRobot, we encourage them to explore the data to its fullest – rather than rushing to train models. Thanks to the time savings we’re seeing with DataRobot, they can take a step back to understand the data in relation to what they’re creating.
This level helps people learn how to operate the DataRobot platform and reveal important information.
At the same time, there is less worry about whether something has been coded correctly. When experts can execute their ideas, they have confidence in what they have created on the platform.
Connecting with a trusted Cloud Computing partner
For cloud computing, we are a pure Amazon Web Services store. By acquiring DataRobot through the AWS marketplace, we were able to start working with the platform within a day or two. If this had lasted a week, as is often the case with new services, we would have had a service outage.
The integration between the DataRobot AI platform and this broader technology ecosystem ensures that we have the infrastructure to effectively address predictive and productive AI initiatives.
Attention to privacy, transparency and accountability
In the highly regulated fintech industry, we must comply with several compliance, security and audit requirements.
DataRobot meets our requirements with transparency, bias mitigation and fairness behind all our models. This helps us to ensure that we are accountable in everything we do.
Standardized workflows Set the stage for continuous innovation
For smoother adoption, creating standard operating procedures was critical. As I experimented with DataRobot, I documented the steps to help my team and others with the integration.
What’s next for us? Data science has changed dramatically in recent years. We make better and faster decisions as AI moves closer to how people behave.
What excites me most about AI is that now it’s essentially an extension of what we’re trying to achieve – like a co-pilot.
Our competitors are probably 10 times bigger than us in terms of team size. With the time we save with DataRobot, we now have the opportunity to move forward. The platform is an extreme developer productivity multiplier that enables our existing experts to prepare for the next generation of engineers and quickly deliver value to our customers.
About the Author
Pranjal Yadav is an accomplished professional with a decade of experience in the technology industry. He currently serves as the Head of AI/ML at Razorpay, where he leads innovative projects that leverage machine learning and artificial intelligence to drive business growth and enhance operational efficiency.
With deep expertise in machine learning, system design and solution architecture, Pranjal has a proven track record of developing and deploying scalable and robust systems. His extensive knowledge of algorithms, combined with his leadership skills, enables him to effectively lead and lead teams, fostering a culture of continuous improvement and excellence.
Throughout his career, Pranjal has demonstrated a strong ability to design and implement strategic solutions that meet complex business requirements. His passion for technology and commitment to development have made him a respected industry leader dedicated to pushing the boundaries of what is possible in the AI/ML space.
Meet Pranjal Yadav