Okay, so let’s get to work!

1. Obtaining their daily loan tape (sometimes with > 1M rows!)

2. Copy/Paste all loans inside the new loan tape into their borrowing base template file

3. Wait until it get processed (sometimes > 1 hour)

4. Check numbers (eligibility criterias, discounts, advance rates, concentration limits, etc.)

5. Add/Update forwards

6. Calculate hedging cost

7. Calculate unrealized gains/losses

8. Put cash reserves

In all those steps, navigating the excel file was very slowly (most of all with heavy files), error prone because all the cell references had to be correct. And the worst part of all: If something was wrong, they had to do it all over again, losing hours of work.

At Vaas, we’ve built a product that receives a loan tape, and the only thing our customer does is to put the cash they need to reserve, the cash available in the trust, and that’s all! In a few clicks and less than a minute it’s all solved.

Nice! And how do you people make sure that your tools do precisely what they are supposed to do?

Personally, I think that’s a really nice and challenging question. Let me try: As some of my colleagues talked about in previous posts, communication is key for us, but also we work really close between departments.

So, when it comes to building solutions, we have our incredible product team documenting all possible use cases in order to test all of them when the product is ready to do so. In fact, testing use cases is one of our last steps in our processes. The next one is to fix or improve anything coming from that stage, and finally deliver the tool.

But that’s not it! We need a lot from our capital markets team. They know the problems better than anyone. So they are the ones who help us to test the harder parts. For instance, some of our tools replace Excel made solutions as we told before, so in order to be really precise we replicate scenarios with known solutions and also, our CM team creates more scenarios with border cases just to make sure we won’t have future problems.

Last but not least, it’s also very important for us to deliver in a safe and reliable way, so these iterations won’t affect what’s already working. We achieve that by creating automated tests that run every time we want to deliver.

If you got this far, you may suspect that we are not really some random guys as I told you before. We take our time to select the best of the best, people who share our culture and passion for solving debt-related problems.

By: Federico Mancovsky | Technical Leader

Other interesting posts