The Wall Street boss was leaning his nose with the idea of open source. Open source is non-dedicated software that developers can view and modify jointly.
Morgan Stanley software developers are used to pushbacks when pitching to build something new.
“I always say to technicians that if you’re building something and not a source of competitive advantage, “Is this already outside? If you’re not using it, do you need to use it? ? Is there a gap where should we improve and contribute to improvement?
“And only then will you create something new.”
Katz talks about prioritizing open source technology, non-specific software that allows all developers, including rival companies, to view, collaborate and modify the code behind it.
The open source-first approach and cooperative attitude are a departure from the ultra-competitive and unique nature of Wall Street, which has become known. But Katz’s stance is part of the industry’s ocean change as it finds reasons why banks, hedge funds and asset management companies are open.
While revealing their secrets, finance companies have abandoned these conservative concepts over the past few years, although alien to Wall Street’s DNA. In that shift, financial institutions are growing not only to consumers but also to be contributors to open source. As a result, the amount of intellectual property flowing to open source hosting platforms and the number of donations from financial services experts continues to grow.
According to the Fintech Open Source Foundation (FINOS), a financial services worker in 2024, the major platform for hosting open source projects, according to the FinTech Open Source Foundation (FINOS), an open source nonprofit organization for financial services, in 2024 financial services workers will be able to host open source projects. Me We have provided over 750,000 commits or changes to GitHub, one of them. This is a 55% increase from the 2021 level. Around the same time, financial companies from JPMorgan to BlackRock and Man Group also released internal platforms for open source.
Wall Street’s open source embrace could be at a good time. According to an October report by McKinsey, bank spending reached $650 billion in 2023. It is almost Belgian or Swedish GDP. Despite a steady 9% increase in spending average annually, bank shareholders and analysts are always questioning the quantifiable value, the report says. Scrutiny occurs when finance companies carve out more resources for AI. This can be a costly effort.
One of the main benefits of leveraging open source is cost-effectiveness. Frees engineers to tackle other tasks while sharing the burden of development. External contributors can implement new features with codebase modifications, allowing businesses to be more agile in the long run and reduce their long-standing roadmap.
“If not, you’re just building up technology debt forever. If you’re building it yourself, you need to support something forever,” Katz said. .
Neema Rafael, chief data officer at Goldman Sachs, has witnessed this cultural change over more than 20 years at the bank.
He said the attitude “from almost very secretive, we’re trying to keep everything up,” and he’s come to accept the benefits of open source with some caution. “Then, in the end, ‘OK, we should be players in this field in the sense that we really need to contribute to open source,” he said.
Katz said there is momentum at the advanced level. He said engineers could work together to solve common industry problems and lead to better solutions. Of course, all the reasons are not altruistic.
Open source reduces development and ownership costs, supports scalability, and even provides regulatory insulation. Banks have now speeded up integration with technology vendors, allowing them to share data with clients more easily. There’s even ripple effect on engineering retention and job satisfaction, Rafael added, “Contributing to open source and having your name in open source is like a table bet.”
Related Stories
The rise of open source on Wall Street
It’s difficult to talk about Wall Street’s open source adoption without talking about the cloud.
Wall Street has primarily embraced the adoption of public clouds, as regulatory uncertainty and security concerns are two major reasons. However, that began to change around 2019 and 2020. The cloud has become a place for data and analysis, a breeding ground for faster experiments and launching new businesses.
To enjoy the benefits of speed and cost-effectiveness, companies had to adopt different toolkits and approaches compared to software development on physical servers. This led many financial companies to face open source technologies such as Kubernetes and Docker, which are the foundations of cloud-native software development.
It is difficult to distill the exact number of financial services workers that have been made on GitHub, but due to company policies and restrictions, developers use personal accounts rather than tied to work emails It will become – Finos’ low ball estimates show a steady increase. The most recent data shows a jump of 26% in commits between 2023 and 2024. Meanwhile, Finos broke over 100 members of the group last month, including four largest US banks, hedge fund Point72, American Express and vendors. Public-Cloud provider and Nvidia.
At the company level, financial companies are also open sourcing platforms originally developed internally. JPMorgan opened sourced its design platform Salt in 2022, while Morgan Stanley opened the morphia platform that helps business and technology teams work together in 2020, and City Open, a risk and control tool that same year I provided the source GitProxy. Other examples include the crown jewels of Goldman Sachs’ major data platforms, Legend, Aladdin and BlackRock technology.
“Ultimately, people are evolving towards the norm, and their competitors are aware of the benefits of being different from the above standard,” Katz said.
Open source is often a software development utility, such as how to write log files, how to connect to a network, formulas for linking datasets, or frameworks for testing applications. Katz described it as “trivial product capabilities” that are not the core of business or promotional values.
“Want to test your application? Your IP is an application. How do you test it? More products,” he said.
Open source hits private equity
Blackstone has a project underway to build an open source library for portfolio companies in the areas of data management, data engineering and generation AI, John Stecher, the company’s chief technology officer, told BI.
They were areas where the PE giants were already spending a significant amount of engineering resources on their own data and AI ambitions.
The goal of the project, which began about a year ago, is to help engineers spend their time on more meaningful work by guiding portfolio companies to a head start in engineering legwork. Blackstone is also profitable as portfolio companies can contribute to their efforts by fixing coding bugs and developing new features and features.
Although Blackstone doesn’t have direct savings, he said portfolio companies should be more effective with fewer engineering resources.
The spirit of repositioning engineering talent is something Stitcher takes priority.
“As we discuss open source, we can move engineers away from building non-dedicated, overlapping features on the street. At the end of the day, there is much higher value for clients,” he said. .
However, open source usage does not occur without risk. According to Goldman’s Raphael, one issue is licensing. Open source software is released under certain licenses with their own requirements, such as disclosure of changes or inability to use for commercial purposes.
“We have completely formal governance on which projects are permitted and the licenses that have been reviewed are permitted.
Another problem is that open source libraries are really available to everyone. It is possible for people to sneak in with bad code that could lead to widespread cyberattacks, what is known as supply chain attacks.
“We take a supply chain of what is actually used and which libraries are used from open source, to ensure that we see where the Providence of all software comes from. and actually track it at a granular level of detail,” Raphael said.
Related Stories
Next Open Source Frontier
Now, finance companies are even thinking about bringing an open source approach to Wall Street’s latest technology frontier AI.
The intersection of generator AI and open source has been fully on display since the Chinese AI lab Deepseek deployed an AI model that is better or even better than Openai’s best products. Deepseek’s offerings rock Silicon Valley and Wall Street, sharply covering Openai with prices. The up-and-coming competition between the two companies has highlighted open source models (like Deepseek) and unique approaches (like Openai).
Last year, Finos launched an AI prep group that attracts companies such as Citi, Morgan Stanley and the London Stock Exchange. This initiative is designed to create a collaborative environment for developing frameworks and policies on generating AI that are relevant to concerns such as mitigating model bias and adopting new technologies.
“It’s a very industry problem, not an individual company problem. It’s better than reciprocalizing it and working together,” Katz said.
