Design atomic Jev questions for reviewable routing
Separate category, evidence sufficiency and impact for fictional product reports, then combine the answers in code. Includes a specification and edge cases.
JEVLOG · FREE PRACTICAL GUIDES
Jev AI guides & tools.
Start with one real problem.
119 guides and project breakdowns organized by real work. Every page states its sources, evidence boundary, and practical outcome.
RECOMMENDED START · SUPPORT TRIAGE
A customer messageJevA reviewable table
Follow one complete path from CSV to human review.
Start with the five questionsFREE · TEACHING DATA · NO AUTO SEND
FREE LOCAL TOOLS
Check your own data before a real decision.
Process CSVs in your browser. No file upload or model call. Try a fictional sample, then download the report.
01 / CSVCheck inputsFind blanks, duplicates, and uneven rows02 / DIFFCompare runsFind changed, missing, and new IDs03 / REVIEWBuild a review queueHold missing and low scores for review
Worked tutorials with inputs and expected reports: CSV preflight ·Run comparison ·Review handoff
Find a guide by practical use
Newest publications first. Search and filters keep the same date order.
NEWEST FIRST · BY PUBLICATION DATE
Choose Jev confidence thresholds with a labeled CSV
Compare automatic coverage and accepted errors across candidate cutoffs, inspect high-confidence mistakes and reserve a validation set. Includes local calculation code.
Write a reviewable Jev Score rubric for issue impact
Replace vague severity labels with observable anchors. Separate impact and evidence, retain fractional scores and human expectations, and produce reviewable priorities.
Handle Jev Python timeouts, retries and error states
Configure bounded SDK retries, distinguish authentication, timeout and response validation errors, retain request IDs and persist task states for refresh and duplicate submissions.
Find faulty CSV rows with DuckDB reject tables
Fix CSV types, collect rejected row locations and reasons with store_rejects, export usable data and errors separately, and reconcile counts with a malformed fixture.
Export PDF tables with Docling and retain provenance
Export candidate PDF tables as CSV and HTML, retain source hashes and table indices, and inspect merged cells, repeated headers and numeric columns using a checklist.
Pause and resume LangGraph approval without duplicate actions
Use interrupt and Command for approval, rejection and resumption. Distinguish thread IDs, in-memory checkpoints and durable storage, with an audit-record checklist.
Check saved outputs with Promptfoo CSV regression fixtures
Use __expected exact assertions and a local JavaScript provider to check saved outputs. An intentional failure verifies reporting without an LLM judge. Includes a complete fixture archive.
Handle n8n failures with error queues and bounded retries
Collect automatic failures with Error Trigger, classify credentials, rate limits and validation, and retain execution IDs and completed steps. Includes an error-register CSV.
Debug browser automation with Playwright traces
Record actions and DOM snapshots on a local fixture, inspect locators and waits in trace.zip, and write a minimal reproduction. Includes a local-page exercise archive.
Build a WeKnora Support Knowledge Base and Check Every Citation
Turn CSV, documents, or search results into reviewable data
Connect LangBot to WeCom Customer Service and test a support bot
Classify support messages, route reviews, and hand off tickets
Source previews load from author media URLs and retain attribution. Offline fallback images are explicitly labeled. Reading a project is not a reproduced test.