Machine Learning powered API governance
summary
Platform has 800 certified apps, handles ~100 million API calls and 2.5 million webhooks monthly; an autonomous system uses clustering analysis on high-level API metadata (endpoint counts, rates) pulled from logs like Splunk/Datadog/Sumo Logic to group partners by use case and flag when a partner’s cluster ID changes. A proof-of-concept reliably surfaces silent pivots (e.g., sudden reads of contact data or inventory posts), targets organizations with ~50+ partners, and takes roughly one to two months to train and tune.
AI-written summary · about
chapters (15)
- 0:00Introduction
- 2:37Cloud Software Business
- 5:15A hidden problem
- 7:12Government platforms rely on manual tasks
- 8:49AI expert system
- 10:51Categorization problem
- 11:24Clustering analysis
- 13:26Do a scenario
- 14:38Standardize
- 15:04Data dump
- 15:55Endpoints data
- 16:15Inventory data
- 17:02End result
- 17:59What kind of clustering analysis
- 18:39QA time
related talks
33mSafeguard Your APIs: Machine Learning Solutions
13mApplying Azure Machine Learning service to drive success of API programs
24mAPI security, mitigate API abuses using machine learning
18mAPIs: ROI from artificial intelligence
21mMonitoring the API Ecosystems at Global Scale
14mAI-Driven API Resilience in Cloud Ecosystems
17mML-Driven API Standardization: Boosting Efficiency & Developer XP
20mAPI Intelligence in an heterogenous world