ParseIt ingests and parses large batches of resumes, extracting entities and ranking candidates using configurable scoring strategies.
Highlights:
- Multi‑format ingestion: PDF, DOCX, image OCR
- Structured extraction: personal info, education, experience, skills
- Configurable scoring: recency, seniority, skill match, tenure
- Bulk processing with parallel workers and deduplication
- Export to CSV/JSON and ATS integrations via webhooks
Architecture:
- API: FastAPI/Node with async workers (RQ/Celery/BullMQ)
- NLP: spaCy/transformers for NER and semantic skill matching
- Queue: Redis/SQS, Storage: S3, DB: Postgres/Mongo
- UI: Next.js dashboard for uploads, search, filters, and exports
Tech stack: Python, FastAPI, spaCy, transformers, Node.js, Next.js, Redis/SQS, PostgreSQL/MongoDB, S3.