Better data quality with ML & visualization
summary
Anomalo monitors an enterprise warehouse with unsupervised machine learning that automatically flags unknown anomalies, routes alerts to Slack/PagerDuty/Teams, and runs root cause analysis that points to exact columns and sample bad rows. Examples shown: an SLA check at 6 AM, a key metric where average tickets moves from ~10 to >13, 233 records failing a 12=3 validation, and a value 40 jumping to >10% of records; visual Time series model* and tornado plots make those issues obvious.
AI-written summary · about
chapters (13)
- 0:00Introduction and setup
- 1:08Why monitor data warehouse anomalies
- 3:57Simplicity in data quality tools
- 8:29Intelligence with unsupervised ML
- 14:30Demo overview and data profiling
- 15:54ML for data arrival and volume checks
- 18:46Unsupervised checks for missing data
- 19:50Table anomalies and drift detection
- 20:54Custom key metrics checks
- 23:25Validation rules for granular checks
- 24:51Root cause analysis feature
- 26:38Automatic anomaly detection demo
- 29:06Summary and Q&A
related talks
28mHow Machine Learning Can Achieve Data Quality Monitoring At Scale
34mML-Bound Anomaly Detection: Experiences Beyond Commodity Observability
12mUnleashing the Power of AI for Performance Monitoring& Anomaly Detection
43mBeyond Logs: Real-Time Anomaly Detection with Machine Learning
24mEmpowering High-Performance Teams with DevOps & Machine Learning
42mMachine Learning In Observability Systems: From Insights To Action
19mThe Rise of AI in Anomaly Detection
26mML Monitoring & Observability