The View from the Office.

I met up with Miguel Jaques, co-founder of startup Terrapin at Pain Quotidien near Columbus Circle. He had a cappuccino. I had lemonade and granola yoghurt parfait.

Terrapin’s story illustrates the tectonic shift fueled first by machine learning and now AI that is reshaping the landscape for financial data.

Terrapin collects reference data for municipal bonds. If you’re not familiar with it, reference data includes the essential details about a security, such as when and by whom a bond was issued.

It is basic, but essential information all the major players require and for years has been provided mainly by a handful of behemoths like Bloomberg, ICE, Standard & Poor’s Financial Services LLC and Refinitiv.

In the past, the data was so time consuming and expensive to gather that it served effectively as a moat around the business. When I started at Bloomberg in 1990, bond data was culled from prospectuses by hand and in some cases that is still how it’s done.

Terrapin is surfing a wave of machine learning that levels the playing field by automatically extracting data at scale. It makes it possible not only to assemble reference data quickly but also do it from anywhere by anyone.

Hence the incongruity of Miguel, who hails from Portugal, teaming up with co-founder Daniel Dyulgerski, CFA, a former Aberdeen fund manager from Bulgaria, to form a company based in Edinburgh, Scotland that focuses on muni bonds, a uniquely American asset class.

Miguel said Terrapin initially collected reference data on corporate bonds but pivoted to munis. In part, that’s because the market is so fragmented, made up of a large number of relatively small issuers. There are more than 50,000 cities and towns that have sold more than 1 million securities.

He said another plus was that municipal bond prospectuses are centralized in a filing repository and tend to be very long — often 300 pages — making them particularly suited to be read by AI.

Miguel is under no illusion that it’s going to be challenging to dislodge large and entrenched competitors.

But he also understands some of the advantages that Terrapin enjoys, such as being less expensive and more nimble.

“There’s always an opportunity for a new guy,” is something I recall Michael Bloomberg saying decades ago when I worked at the company.

I know Miguel would agree.

You can reach Miguel via LinkedIn or DM me for a warm intro. hashtag#viewfromoffice