A Claude Skill to render resume templates. CV/Resumes are HTML and JSON
JSON content plus Jinja2 templates solves the AI chat one-off resume problem.

Schema-controlled extraction when Rossum and Hyperscience already exist.
Developers building document processing pipelines, HR and fintech teams
Rossum · Hyperscience · AWS Textract
JSON content plus Jinja2 templates solves the AI chat one-off resume problem.
Append-only parsing with delta-only DOM updates solves the 'wait for complete JSON' problem cleanly.
Beats PyPDF and MarkItDown on accuracy without needing GPUs or cloud APIs.
Replaces O(n²) token re-parsing with true O(n) streaming; Vercel SDK does 4K re-parses on 50KB payloads.
424k statements/sec with zero dependencies — sqlparse can't match this performance.
Email-to-JSON with schema validation and webhook delivery, but LLM extraction isn't novel.