recommendation-system
Deploy production recommendation systems with feature stores, caching, A/B testing. Use for personalization APIs, low latency serving, or encountering cache invalidation, experiment tracking, quality monitoring issues.
92
92%
Does it follow best practices?
Impact
92%
1.21xAverage score across 3 eval scenarios
Advisory
Suggest reviewing before use
Evaluation results
97%
19%Recommendation Caching Service
Multi-tier caching and cold start
Redis MGET batch fetch
100%
100%
L1 in-memory tier
100%
100%
L2 Redis tier
100%
100%
Recommendation cache key format
50%
100%
Feature cache key format
0%
100%
Recommendation TTL ~300s
25%
100%
Feature TTL ~3600s
100%
62%
TTL jitter
100%
100%
Cold start fallback
100%
100%
Cache invalidation on actions
60%
100%
L1 promotion from L2
100%
100%
100%
16%Recommendation Algorithm Experimentation Framework
A/B testing framework
MD5 deterministic assignment
41%
100%
Sample size calculation
100%
100%
Thompson Sampling implementation
100%
100%
Thompson Sampling update
87%
100%
Bayesian significance test
70%
100%
Frequentist significance test
100%
100%
SQLite assignment log
90%
100%
SQLite event log
60%
100%
Traffic allocation validation
100%
100%
Variant results query
100%
100%
80%
13%Recommendation Service Observability
Prometheus monitoring and alerts
prometheus-client Counter
100%
100%
prometheus-client Histogram
100%
100%
prometheus-client Gauge
100%
100%
/metrics endpoint
100%
100%
Diversity score tracking
50%
100%
Catalog coverage tracking
100%
100%
CTR alert threshold
0%
50%
Latency alert threshold
100%
50%
Error rate alert threshold
0%
0%
FastAPI integration
100%
100%
model_version label
0%
100%
- Repository
- secondsky/claude-skills
- Commit
90d6bd7
- Evaluated
- Agent
- Claude Code
- Model
- Claude Sonnet 4.6
Table of Contents
Is this your skill?
If you maintain this skill, you can claim it as your own. Once claimed, you can manage eval scenarios, bundle related skills, attach documentation or rules, and ensure cross-agent compatibility.