A Unified Architecture for Real-Time Decision Engines and Customer Analytics in Banking
DOI:
https://doi.org/10.68337/cpsm.v1.i2.2026-026Keywords:
Real-time decisioning, banking analytics, data lakehouse, feature store, MLOps, SR 11-7, change data capture, agentic AI, champion-challenger, closed-loop learningAbstract
Banks hold plenty of customer information, yet they still base many decisions on data that are hours or even days old. This paper addresses the problem with the Real-Time Analytic Decisioning (RTAD) architecture, which fuses four elements: a change data capture (CDC) streaming pipeline, a Delta lakehouse, a centralized feature store, and a machine learning (ML) decisioning engine governed under Federal Reserve SR 11-7. Drawing on documented transformations at Wells Fargo and PNC Financial Services, and on benchmarks from a purpose-built reference implementation, the paper reports an end-to-end CDC latency of 4.9 s (P95), online feature retrieval in 8 ms (P99), a 3.3× improvement in decisioning round-trip time, and a 38% reduction in per-decision compute cost over a daily-batch baseline. It also describes how the same substrate supports agentic artificial intelligence (AI) workflows under banking-grade safety constraints without duplicating the regulatory infrastructure already in place.
References
[1] Board of Governors of the Federal Reserve System, SR Letter 11-7: Guidance on Model Risk Management. Washington, DC, USA: Federal Reserve, 2011. [Online]. Available: https://www.federalreserve.gov/supervisionreg/srletters/sr1107.htm
[2] Michael Stonebraker and Uğur Çetintemel, "'One size fits all': An idea whose time has come and gone," in Proc. 21st Int. Conf. Data Eng. (ICDE'05), 2005, pp. 2-11, doi: 10.1109/ICDE.2005.1.
[3] Michael Armbrust, Ali Ghodsi, Reynold Xin, and Matei Zaharia, "Lakehouse: A new generation of open platforms that unify data warehousing and advanced analytics," in Proc. Conf. Innov. Data Syst. Res. (CIDR), 2021.
[4] Michael Armbrust, Tathagata Das, Liwen Sun, Burak Yavuz, Shixiong Zhu, Mukul Murthy, Joseph Torres, Herman van Hovell, Adrian Ionescu, Alicja Łuszczak, Michał Świtakowski, Michał Szafrański, Xiao Li, Takuya Ueshin, Mostafa Mokhtar, Peter Boncz, Ali Ghodsi, Sameer Paranjpye, Pieter Senster, Reynold Xin, and Matei Zaharia, "Delta Lake: High-performance ACID table storage over cloud object stores," Proc. VLDB Endow., vol. 13, no. 12, pp. 3411-3424, 2020, doi: 10.14778/3415478.3415560.
[5] Jay Kreps, Neha Narkhede, and Jun Rao, "Kafka: A distributed messaging system for log processing," in Proc. 6th Int. Workshop Netw. Meets Databases (NetDB), 2011.
[6] Tyler Akidau, Robert Bradshaw, Craig Chambers, Slava Chernyak, Rafael J. Fernández-Moctezuma, Reuven Lax, Sam McVeety, Daniel Mills, Frances Perry, Eric Schmidt, and Sam Whittle, "The dataflow model: A practical approach to balancing correctness, latency, and cost in massive-scale, unbounded, out-of-order data processing," Proc. VLDB Endow., vol. 8, no. 12, pp. 1792-1803, 2015, doi: 10.14778/2824032.2824076.
[7] D. Sculley, Gary Holt, Daniel Golovin, Eugene Davydov, Todd Phillips, Dietmar Ebner, Vinay Chaudhary, Michael Young, Jean-François Crespo, and Dan Dennison, "Hidden technical debt in machine learning systems," in Adv. Neural Inf. Process. Syst., vol. 28, 2015, pp. 2503-2511.
[8] Matei Zaharia, Andrew Chen, Aaron Davidson, Ali Ghodsi, Sue Ann Hong, Andy Konwinski, Siddharth Murching, Tomas Nykodym, Paul Ogilvie, Mani Parkhe, Fen Xie, and Corey Zumar, "Accelerating the machine learning lifecycle with MLflow," IEEE Data Eng. Bull., vol. 41, no. 4, pp. 39-45, 2018.
[9] Scott M. Lundberg and Su-In Lee, "A unified approach to interpreting model predictions," in Adv. Neural Inf. Process. Syst., vol. 30, 2017, pp. 4765-4774, doi: 10.48550/arXiv.1705.07874.
[10] Agus Sudjianto, Jinwen Qiu, Miaoqi Li, and Jie Chen, "Linear iterative feature embedding: An ensemble framework for interpretable model," arXiv:2103.09983, 2021, doi: 10.48550/arXiv.2103.09983.
[11] Patrick Lewis, Ethan Perez, Aleksandra Piktus, Fabio Petroni, Vladimir Karpukhin, Naman Goyal, Heinrich Küttler, Mike Lewis, Wen-tau Yih, Tim Rocktäschel, Sebastian Riedel, and Douwe Kiela, "Retrieval-augmented generation for knowledge-intensive NLP tasks," in Adv. Neural Inf. Process. Syst., vol. 33, 2020, pp. 9459-9474, doi: 10.48550/arXiv.2005.11401.
[12] Tom B. Brown, Benjamin Mann, Nick Ryder, Melanie Subbiah, Jared Kaplan, Prafulla Dhariwal, Arvind Neelakantan, Pranav Shyam, Girish Sastry, Amanda Askell, Sandhini Agarwal, Ariel Herbert-Voss, Gretchen Krueger, Tom Henighan, Rewon Child, Aditya Ramesh, Daniel M. Ziegler, Jeffrey Wu, Clemens Winter, Christopher Hesse, Mark Chen, Eric Sigler, Mateusz Litwin, Scott Gray, Benjamin Chess, Jack Clark, Christopher Berner, Sam McCandlish, Alec Radford, Ilya Sutskever, and Dario Amodei, "Language models are few-shot learners," in Adv. Neural Inf. Process. Syst., vol. 33, 2020, pp. 1877-1901, doi: 10.48550/arXiv.2005.14165.
[13] Consumer Financial Protection Bureau, Required Rulemaking on Personal Financial Data Rights (Section 1033 of the Dodd-Frank Act), Final Rule, 12 C.F.R. Part 1033, 2024.
Downloads
Published
Data Availability Statement
The data supporting the findings of this study are available from the corresponding author upon reasonable request.Conference Proceedings Volume
Section
Categories
License
Copyright (c) 2026 The Authors
This work is licensed under a Creative Commons Attribution 4.0 International License.
Articles published in Conference Proceedings in Science and Management are made available under the Creative Commons Attribution 4.0 International License (CC BY 4.0). Authors retain copyright in their work and grant the journal first publication rights. Readers are free to share and adapt the published material for any purpose, including commercial use, provided appropriate credit is given to the author, a link to the license is provided, and any changes are indicated.
