The FINOS TechSprint is an extended collaborative effort that invites developers to explore cutting-edge financial solutions using open source tools. The primary focus of the sprint is TraderX, a highly modular, open source trading platform designed to serve as a reference architecture for integrating various FINOS projects. This hackathon-style event runs through September, culminating in demonstrations at the Open Source in Finance Forum (OSFF) New York. Haven’t registered yet for OSFF NYC? Use the code OSFFNY24SOC to get 20% off your ticket and register here.
FINOS continues to break new ground with its TechSprint 2024, an ongoing initiative to foster open source collaboration in the financial services industry. On September 12, developers gathered at the Microsoft Times Square office to engage in hands-on coding, aiming to address key challenges in financial services through open innovation. The event brought together different teams, each hacking away at their own workstreams, but ultimately contributing to the broader FINOS ecosystem.
Key Achievements from the TechSprint so far
- Identifying Climate Risks in Mortgages: One workstream is focusing on analyzing the impact of climate change on residential mortgages. The team uses OS-Climate tools to create new analytics that integrate climate scenarios like floods and wildfires into the valuation models for mortgage portfolios. This data is essential to stress-test these assets, offering developers the chance to create meaningful insights into future risks. Given that the sprint centered around TraderX, much of the work is being devoted to enhancing the platform’s ability to handle US Mortgage Backed Securities (MBS) trading, the most liquid market for this asset class.
As a next step, participants will be exploring how traditional financial risks such as prepayment rates (CPR) and default rates (CDR) could be complemented by climate risk factors using LEGEND, a Data Modeling Platform contributed by Goldman Sachs into FINOS.
- Open Source Innovation with CDM and FinGPT: Tom Healey (ICMA) enhanced the TraderX Data Processor service to be compatible with FINOS Common Domain Model (CDM). CDM, a standard initially created by ISDA (International Swaps and Derivatives Association), is designed to streamline the trading and post-trading processes by providing a standardized representation of financial products. With this update, TraderX can now support any asset class covered by the CDM standard, enabling the platform to potentially handle a wider array of financial instruments within a unified framework.
At the upcoming OSFF NYC event, during the “AI Meets CDM” workshop, the FINOS LLM Exploration will present their findings on implementing retrieval-augmented generation (RAG) techniques for the CDM documentation. Attendees will have the opportunity to engage directly with live chatbot demonstrations, asking questions, evaluating responses, and experiencing an innovative approach to interactive learning.
This enhancement significantly accelerates the process of mapping new asset classes to CDM. If scaled to “enterprise-level”, this could enable the mapping of virtually any asset class globally into the CDM standard. One major outcome of this open-source innovation related to the TechSprint is the ability to extend CDM to include new asset classes like mortgage-backed securities (MBS), thus expanding TraderX’s capabilities in handling complex financial products.
- Bi-Temporal Data for Accurate Historical Analytics: In a separate workstream, developers enhanced TraderX by integrating XTDB, a bi-temporal database, allowing for more nuanced historical analysis of trades. This upgrade supports better analytics and ensures that data can be accurately tracked across multiple dimensions. Learn more in the dedicated GitHub Discussion.
- Radius Deployment: Cloud-Native Scalability for TraderX: One of the exciting use cases involved deploying TraderX using Radius, a cloud-native platform designed for scalability and portability. This migration allows the trading platform to efficiently scale components, such as databases and message bus resources, in a distributed environment. The team worked on automating the end-to-end CI/CD pipeline to ensure seamless deployment and integration. Ler more in the dedicated GitHub Discussion
Trading for Consumers: A final workstream focused on connecting a retail trading app to TraderX, making trading of multiple asset classes more accessible for consumers, Learn more in the dedicated GitHub Discussion.
See you next week at OSFF NYC?
A live demo will be showcased at the OSFF NYC TraderX Booth, where participants can witness how Radius, XTDB and other technologies enhance TraderX’s flexibility and support rapid experimentation and testing in a cloud environment.
The FINOS TechSprint embodies the best of hackathon culture - rapid iteration, collaborative coding, and open innovation. As we move towards the OSFF New York Demo, it is clear that the solutions developed during this sprint could have significant, real-world implications for financial services.
Through open source collaboration, FINOS is enabling the financial industry to adapt to emerging risks, such as climate change, and embrace the power of innovation.
.
JOIN US AT THE OPEN SOURCE IN FINANCE FORUM (OSFF) THIS YEAR