83% Faster, 10x Fewer Errors: Engineering a Self-Healing Payment Saga Tracker
Abstract
This paper describes the re-architecture of a distributed payment system around an event-driven saga pattern, tracking transactions from initiation through settlement and automatically triggering compensations when a step fails. The results are given directly: transaction processing time fell by 83 per cent, payment errors dropped tenfold, and security incidents fell by 95 per cent. The pattern's value is that it converts fragile, error-prone distributed transactions into systems that recover without intervention, which matters most in payments, where a partial failure left uncompensated is a financial discrepancy rather than a technical one. The paper also covers a companion project, an Azure OpenAI-powered chatbot using retrieval-augmented generation and vector embeddings, which now handles between 50 and 70 per cent of customer support queries and reduced support costs by 10 to 15 per cent. Taken together, the two show how reliability engineering and AI-powered support compound, improving operational efficiency and cost at the same time.
- Published in
- Proceedings of the Raptors Conference, Vol. 1 (2026)
- Conference
- Raptors Conference 2026, 21 October 2026, online
- Accepted
- 2026-09-19
- DOI
- https://doi.org/10.68495/rc.2026.C074
- Presentation
- Slides (PDF)
- Licence
- © 2026Tarun Kumar Chatterjee. Published under Creative Commons Attribution 4.0 International (CC BY 4.0).
- Publisher
- Hackathon Raptors, Community Interest Company 15557917, London, United Kingdom
- Permanent address
- https://proceedings.raptors.dev/chatterjee-83-faster-10x-fewer
Open access. Every entry in the Proceedings is free to read, with no charge to authors and no charge to readers. Slides are served from this site, so they stay available at the paper's permanent address.
Cite this paper
Tarun Kumar Chatterjee (2026). 83% Faster, 10x Fewer Errors: Engineering a Self-Healing Payment Saga Tracker. Proceedings of the Raptors Conference, Vol. 1 (2026). Hackathon Raptors, London. https://doi.org/10.68495/rc.2026.C074
@inproceedings{raptors2026-c074,
title = {{83\% Faster, 10x Fewer Errors: Engineering a Self-Healing Payment Saga Tracker}},
author = {Chatterjee, Tarun Kumar},
booktitle = {Proceedings of the Raptors Conference},
volume = {1},
year = {2026},
publisher = {Hackathon Raptors},
address = {London, United Kingdom},
doi = {10.68495/rc.2026.C074},
url = {https://doi.org/10.68495/rc.2026.C074},
abstract = {This paper describes the re-architecture of a distributed payment system around an event-driven saga pattern, tracking transactions from initiation through settlement and automatically triggering compensations when a step fails. The results are given directly: transaction processing time fell by 83 per cent, payment errors dropped tenfold, and security incidents fell by 95 per cent. The pattern's value is that it converts fragile, error-prone distributed transactions into systems that recover without intervention, which matters most in payments, where a partial failure left uncompensated is a financial discrepancy rather than a technical one. The paper also covers a companion project, an Azure OpenAI-powered chatbot using retrieval-augmented generation and vector embeddings, which now handles between 50 and 70 per cent of customer support queries and reduced support costs by 10 to 15 per cent. Taken together, the two show how reliability engineering and AI-powered support compound, improving operational efficiency and cost at the same time.}
}
TY - CONF TI - 83% Faster, 10x Fewer Errors: Engineering a Self-Healing Payment Saga Tracker AU - Chatterjee, Tarun Kumar T2 - Proceedings of the Raptors Conference VL - 1 PY - 2026 DA - 2026/09/19 PB - Hackathon Raptors CY - London, United Kingdom DO - 10.68495/rc.2026.C074 UR - https://doi.org/10.68495/rc.2026.C074 AB - This paper describes the re-architecture of a distributed payment system around an event-driven saga pattern, tracking transactions from initiation through settlement and automatically triggering compensations when a step fails. The results are given directly: transaction processing time fell by 83 per cent, payment errors dropped tenfold, and security incidents fell by 95 per cent. The pattern's value is that it converts fragile, error-prone distributed transactions into systems that recover without intervention, which matters most in payments, where a partial failure left uncompensated is a financial discrepancy rather than a technical one. The paper also covers a companion project, an Azure OpenAI-powered chatbot using retrieval-augmented generation and vector embeddings, which now handles between 50 and 70 per cent of customer support queries and reduced support costs by 10 to 15 per cent. Taken together, the two show how reliability engineering and AI-powered support compound, improving operational efficiency and cost at the same time. LA - en ER -