How BlockFi Is Using Machine Learning To Take Crypto Safety to the Moon!

Anthony G. Tellez2 min read
Machine LearningSecurityCryptocurrencyBlockchainGraph AnalyticsBlockFiConference

BlockFi is a cryptocurrency platform that allows its clients to grow wealth through various financial products capabilities, including loans, trading, and interest accounts.

The safety of our client's assets and personal information is taken very seriously by the security operations team.

What You'll Learn

This session showcases how BlockFi uses Splunk to identify operational risks and ensure the safety of client assets leveraging different machine learning techniques.

Machine Learning Techniques

Anomaly Detection

  • Identifying unusual transaction patterns
  • Detecting account takeover attempts
  • Monitoring for suspicious behavior

Forecasting

  • Predicting potential security threats
  • Capacity planning for infrastructure
  • Risk assessment modeling

Graph Analytics

  • Data mining blockchain transactions
  • Tracing fund flows across wallets
  • Identifying connected threat actors

Use Cases

Fraud Identification

Machine learning models to detect fraudulent transactions and activities in real-time.

Account Takeover Prevention

Behavioral analytics to identify and prevent unauthorized account access.

Blockchain Analysis

Using graph theory to analyze blockchain data and ensure client safety through:

  • Transaction tracing
  • Address clustering
  • Risk scoring

ML Operations

Improving machine learning operations through:

  • Summarization - Efficient data aggregation for model training
  • Feature Development - Engineering meaningful features from crypto data
  • Model Deployment - Operationalizing models in production environments

Presentation Materials

Download Slides


Presented at Splunk .conf21