How I Built a Real-Time Fraud Detection System That Handles 71,000 RPS at p95 <6ms
The article discusses the development of Sentinel, a real-time fraud detection system capable of processing 71,000 requests per second with a response time of under 6 milliseconds. It highlights the challenges of fraud detection, including the need for real-time classification and high accuracy. The author shares insights on using XGBoost and ONNX for model training and inference, emphasizing the performance benefits of implementing the system in Go.
- ▪Sentinel processes 7.8 million requests with zero errors using machine learning techniques.
- ▪The system was designed to handle high throughput while maintaining accuracy and minimizing downtime.
- ▪The author trained the model on a heavily imbalanced dataset and achieved a PR-AUC of 0.87.
Opening excerpt (first ~120 words) tap to expand
try { if(localStorage) { let currentUser = localStorage.getItem('current_user'); if (currentUser) { currentUser = JSON.parse(currentUser); if (currentUser.id === 3965551) { document.getElementById('article-show-container').classList.add('current-user-is-article-author'); } } } } catch (e) { console.error(e); } Sameer Ahmed Posted on Jun 3 How I Built a Real-Time Fraud Detection System That Handles 71,000 RPS at p95 <6ms #go #distributedsystems #machinelearning #programming How I Built a Real-Time Fraud Detection System That Handles 71,000 RPS at p95 <6ms A deep dive into building Sentinel — an ML inference pipeline that processes 7.8M requests with zero errors, using XGBoost, ONNX, and Go. The Problem Fraud detection is a classic hard problem in systems design.
…
Excerpt limited to ~120 words for fair-use compliance. The full article is at DEV.to (Top).
Discussion
0 commentsMore from DEV.to (Top)
