Agent skill · data analytics · jeremylongshore
juicebox-performance-tuning
Optimize Juicebox performance.
Why this skill is useful
Provides specific caching, batching, and error handling strategies that enhance performance for Juicebox's API, which the AI wouldn't reliably generate on its own.
What it needs
About 3k tokens when loaded. Last updated 2026-08-07. 2,604 stars on the source repository.
What this skill does
Juicebox Performance Tuning Overview Juicebox's AI analysis API handles dataset uploads, analysis queue wait times, and result pagination. Large dataset uploads (100K+ rows) can block the analysis pipeline, while queue contention during peak hours increases wait times. Result sets from broad queries return thousands of profiles requiring efficient pagination. Caching search results, batching enrichment calls, and managing upload chunking reduces end-to-end analysis time by 40-60% and keeps interactive searches responsive. Caching Strategy Batch Operations Connection Pooling Rate Limit Management Monitoring Performance Checklist [ ] Use specific filters (location, skills, title) to narrow search scope [ ] Cache search results with 5-min TTL to avoid redundant queries [ ] Batch profile enrichment in groups of 50 with 300ms delays [ ] Chunk large dataset uploads into 10K-row segments [ ] Cache analysis results for 15 min (expensive to recompute) [ ] Set 60s timeout for upload and analysis endpoints [ ] Monitor queue wait times and schedule uploads during off-peak [ ] Paginate results with limit=20 and cursor for interactive UIs Error Handling Issue Cause Fix ------- ------- ----- Analysis queue timeout Peak hour contention Schedule large analyses off-peak, increase client timeout 429 on bulk enrichment Too many concurrent enrichment calls Batch to 50 profiles with 300ms interval Upload failure on large dataset Payload exceeds limit or connection drop Chunk into 10K-row segments, retry failed chunks Slow broad search Unfiltered query returning thousands of results Add location/skills/title filters, set limit=20 Prerequisites An approved performance baseline, synthetic sandbox fixture, bounded test budget, source/destination allowlists, suppression controls, and a rollback owner. Instructions 1. Benchmark cache, batching, and pagination changes against synthetic fixtures only; reject live contact export and unapproved destinations. 2. …
How to use it
Reference it in AdaL, Claude Code, Cursor or any coding agent — nothing to install:
@skills jeremylongshore/juicebox-performance-tuning--b24ebd