Syed Mehdi.000
Syed Mehdi
← All projects
AI

Fraud Detection with LSTM Autoencoder

Real-time anomaly detection on financial transactions, deployed serverless on AWS Lambda.

How it works
Result
+20% recall
Data flows left to right · click a stage
01 Transactions
Streaming input

Live financial transactions flow in as time-ordered events for real-time scoring.

+20%
Fraud recall
-35%
False positives
Real-time
Scoring
Stack
PyTorchLSTM AutoencoderAWS LambdaDockerCI/CD
The Problem

Fraud is rare and ever-changing, so labelled data is scarce and rule-based systems miss novel patterns.

Objective

Detect anomalous transactions in real time with high recall and few false positives.

Approach

Engineered a deep LSTM autoencoder that learns normal transaction sequences; high reconstruction error flags anomalies. Containerised and deployed on AWS Lambda for live scoring, with a CI/CD pipeline for model versioning and automated redeployment.

Challenges

Tuning the reconstruction-error threshold to balance recall against false positives on highly imbalanced data.

Results

Improved fraud recall by 20% and reduced false positives by 35%.

What's Next

Online learning to adapt thresholds as fraud patterns drift.

Want something similar built?

Start a conversation
HomeProjectsServicesResearchWriting