11.21.2019

Goldman Sachs Open Sources Data Modeling

11.21.2019

Goldman Sachs has made another open-source contribution with the donation of its visual model tool Alloy and Pure logical modeling language to the Fintech Open Source Foundation.

“Alloy lets you design, build, and publish data pipelines,” said Neema Raphael, co-chief data officer at Goldman Sachs, at Open Source Strategy Forum hosted by Finos in Midtown Manhattan. “It’s a platform that lets you run those reliably at a quite high volume and as a hosted service.”

If a firm wanted to share data, such as trade data, with broader audiences internally or externally, Alloy lets users take various data attributes, define them as business processes, normalize the data, and publish the normalized data via APIs, he added.

Pure-based models should reduce system integration costs in bilateral and multilateral trading scenarios as well as lower compliance burned and complexity for banks and regulators, noted Gabriele Columbro, executive director at Finos, in a prepared statement.

Goldman Sachs and Finos have had an ongoing discussion regarding the Alloy and Pure donations for the past six months, Rob Underwood, director of programs at Finos, told Markets Media.

Finos leadership learned that its Financial Object Program, which is led by J.P. Morgan and Citi, was looking for a visual modeler to help the project’s object modeling.

“It was then when we first started chatting with Goldman Sachs about its interest in contributing Pure and Alloy,” he said. “It made a ton of sense since we already knew that there was a demand.”

The industry body plans to roll out the new code in three phases. In the first phase, which is taking place currently, FINOS has set up a testing sandbox while Goldman Sachs decouples the code from other systems within the investment bank and adds in a bit more abstraction into the code.

“The sandbox is pretty much ready at this point,” said Underwood. “It uses GitLab on the backend to do the source control of the models.”

In the next phase, which is slated to start in January 2020, Finos will invite organizations to get their feet wet with the modeling code.

Finos expects the final phase, in which it makes the code available to everyone with an internet connection, sometime around mid-2020.

For those that want to contribute code back to Alloy and Pure, they will need an executed contributor license agreement, which is common in the open-source ecosystem, according to Underwood. “As long as people have executed contributor license agreements, they can contribute code for consideration to be included back into the project.”

Pension funds, sovereign wealth funds, endowments and other institutional asset owners are sitting on vast troves of data -- but extracting value from that data is more challenging than ever.

#AssetOwners #DataQuality

Technology costs in asset management have grown disproportionately, but McKinsey research finds the increased spending hasn’t consistently translated into higher productivity.
#AI #Fiance

We're in the FINAL WEEK for the European Women in Finance Awards nominations – don't miss your chance to spotlight the incredible women driving change in finance!
#WomenInFinance #FinanceAwards #FinanceCommunity #EuropeanFinance @WomeninFinanceM

ICYMI: @marketsmedia sat down with EDXM CEO Tony Acuña-Rohter to discuss the launch of EDXM International’s perpetual futures platform in Singapore and what it means for institutional crypto trading.
Read the full interview: https://bit.ly/45xRUWh

Load More

Related articles

  1. The typology will help trading firms ready themselves for the pending European consolidated tape.

  2. This enables traders to anticipate volatility, minimize market impact & optimize execution in real time.

  3. This is a significant milestone towards mainstream adoption of onchain finance.

  4. From The Markets

    SIX Selects Corvil Analytics

    Corvil Analytics provides improved data transparency and helps optimize low-latency data delivery.

  5. The processes used by exchanges to manage market data relationships are becoming increasingly critical.

We're Enhancing Your Experience with Smart Technology

We've updated our Terms & Conditions and Privacy Policy to introduce AI tools that will personalize your content, improve our market analysis, and deliver more relevant insights.These changes take effect on Aug 25, 2025.
Your data remains protected—we're simply using smart technology to serve you better. [Review Full Terms] | [Review Privacy Policy] By continuing to use our services after Aug 25, 2025, you agree to these updates.

Close the CTA