Return to page


AI in the Financial Industry: 8 Key Takeaways from the + Fireside Chat


By Bruna Smith | minute read | November 05, 2020

Blog decorative banner image

The current global pandemic crisis presents various challenges to businesses in all industries, including financial services institutions, who are monitoring and dealing with the effects of COVID-19 across the world. At a time of a pandemic, it is important that teams get together to share their insights and experience, with the goal of inspiring and helping to contain and mitigate the negative impact.

With this in mind, hosted a virtual fireside chat with , where business leaders from both companies discussed the latest in AI, key factors to building and recruiting an AI-first engineering team, the impact of COVID-19 for SMBs, and the lessons learned from the pandemic in the finance industry.

Check out the key takeaways from Sri Ambati’s conversation with Vinay Pai, SVP of Engineering at , Manasa Murthy, VP of Engineering at , and Prakruthi Narasimha, Senior Accountant at, on how to thrive during challenging times and the role of AI in shaping the future for the best.

1. The State of SMB 

Sri Ambati kicked off the fireside chat by offering a deeper look into the state of Small and Medium Business (SMB) with data gathered from a Salesforce Research report that analyzed the responses of more than 2,300 small and medium business (SMB) owners and leaders around the world to determine the impact of the COVID-19 pandemic, the role of digital transformation in terms of enhancing business resiliency, and how SMB leaders are planning for recovery and growth post-pandemic.

According to the report , the world of SMBs radically changed in early 2020 due to the COVID-19 pandemic. With a significant reduction in revenue, most SMB leaders say they are struggling to keep their businesses afloat, and businesses that the consumed-focused and micro-businesses were the most affected. Also, besides the challenge of getting access to capital and cash-flow, SMBs are seeing a reduced customer demand, supply chain disruptions, and challenges with public health mandates.

2. Bringing teams together in times of COVID-19 

Vinay Pai:  One thing that I feel really lucky about is that being in tech we have the kind of work that we can do remotely. So we took the whole company remote at the beginning of March. And we started with four principles: stay healthy and safe, stay connected to your teammates, keep running for customers, and then focus on outcomes. We actually did a week-long innovation week, which was a hackathon, so we’re making this work, we’re growing the business. And even since March, we probably hired about 30 engineers.

Manasa Murthy:  I think a lot of the challenges all of us feel going remote with all the uncertainties of 2020 that are presented to us, it’s a lot about embracing change. We need to look at it wholeheartedly. We bring our whole selves to work. We all have kids running around during conference calls. We all have to tend to probably person things during the day. I think it’s really about listening to what’s working and what’s not.

3. Building a gender-balanced team 

Vinay Pai:  Our CEO, Rene Lacerte, believes in diversity and inclusion and it’s reflected in the culture. If our teams reflect the same composition as our customers, you end up making better products. Creating a more diverse and inclusive culture is not something that you solve right away. It’s a long game.

4. The AI journey at 

Vinay Pai:  This goes back to three years ago when I joined, during that second generation of AI at My engineering team had already invested in building some of our own models and, as you know, we get a lot of invoices that our customers upload. So the challenge we had was: how do we extract data from invoices, bring them into the system to minimize data entry? We also do risk and fraud.

I started using this term “democratizing data science ”, and it’s like, how do you take the data we have and make it accessible to everyone in the company, whether you’re in product management or marketing or sales or analytics, where there’s so much we can learn from the data? So this was kind of a two or three-step process.

One is, we started building our own ML models around invoices, risk, and fraud. We ran into at Money2020, we looked at you guys and one of your competitors, and we did a bake-off internally, you guys won. And then we decided to get access to some early adopters and product management that were very data-driven and get them trained along with our engineers, on Driverless AI . And let’s expand that out at the same time. I wanted to build our own AI capabilities.

Part of that was hiring Manasa to lead the team and the next day hiring several data scientists. It’s been a very rich collaboration. Then, with the toolset, we brought H2O to a bunch of folks in the company that they’re using it in their day-to-day lives, which is very powerful.

Manasa Murthy:  If you think about a product like ours, where there’s so much opportunity to apply AI use cases all across the board, I feel like bringing H2O was really useful in accelerating our business use cases. Model building went from months to days, which is a very fast development. We’ve also been able to expand to so many new use cases. Early on when our team was smaller, it was incredibly useful to have a tool like this to expand, and while we build expertise.

Sri Ambati:  Our core vision is to democratize AI. But of course, we pride ourselves in trying to make our customers the AI companies.

5. AI and helping customers and end-users in their day-to-day responsibilities 

Prakruthi Narasimha:  I’ve been using for about two to three years now, very much so on a regular basis. The whole AI functionality was such a game-changer from an end-user perspective. The fact that most of the information was coded and the bill that was being uploaded, it could be a PDF, or word, or an Excel file, or it could be just something that I took a picture from my phone. It would still capture 90% of the important information that I need. In terms of data entry, I’m doing little to no data entry. I’m just verifying information. That gives me a lot of confidence in the accuracy of the bills being entered.

But there are also a lot of other things that offers that set it apart from a lot of AP companies. Such as keeping the whole end-to-end AP on the cloud, and giving us the ability to do the job remotely, without having the need to be in a physical space. All of this without compromising on things like segregation of duties, and having a good approval integrated workflow.

6. Recommendations on the business value of AI  

Vinay Pai : I think AI is one of these things that gets a lot of bad press. Because there’s a lot of use cases where it’s being misused or used inappropriately and those are the things that get on the news. But then if you look at AI and the problems we’re solving. So I feel like in the pandemic and the recession, the fact that we can help businesses manage their business remotely, continue to pay their vendors, continue to get paid by their customers, it’s a great mission where we can help them manage their business and actually grow their business. So it’s coming from a place of solving real customer problems.

Manasa Murthy:  Where I have seen success is when you start from the customer problem. Then it becomes a real issue that you can see the benefit very clearly. Otherwise, AI is just a buzzword; it’s a tech play, “Why do I care?” And AI, especially having such a high initial cost, you really need to start thinking about how do you actually balance that startup boosted cost versus the benefit that you’re going to get overtime. My recommendation would be to focus on value and focus on gathering the data and foundation first.

7. Building an AI-first engineering team  

Vinay Pai:  I think the biggest challenge of any leader is building the team that’s going to, first of all, deliver what you have today, what you need to deliver today, as well as build for the future, and help grow and scale. So you’re always trying to bring in people smarter than you because you want to average upon the IQ of the team and bring in the skill sets. But at the same time, you want to bring along the team that’s there, because they have the scars and the battle stories, and they know the customer base. So it’s a fun engineering problem. So that’s really the factors on how we recruit and build.

8. The Future of AI  

Sri Ambati:  In the next few years of innovation, thanks to AI, now being part of discovery itself, we’ll see such dramatic acceleration in innovation because the time to do an experiment and the cost to do an experiment has been dramatically reduced. When we say democratize AI, we’re really meaning faster, cheaper, easier ways to do experiments and making the experience so inexpensive that it’s almost natural to fail and learn from the failures and then quickly reinvent. But in general, I think being human is being redefined because AI is just making things much more intelligent, and as things become intelligent we can focus on being more human. In general, I think you will probably see a lot more empathy and gratitude.

Click here to watch the recording of the panel. 


Bruna Smith

Bruna is a Field Marketing Manager for Latin America at She is a passionate professional with 10 years of experience, ranging from Internal/Corporate Communication to Marketing and Social Media. Prior to joining, Bruna worked as a Senior Communication Analyst for several years at the largest telco in Brazil and one of the top 3 in Latin America. Bruna holds a Master’s Degree in Strategic Communication at the University of San Francisco (USF) and a Bachelor’s Degree in Social Communication at PUC-Rio.