- Home
- AI Recruiting Assistants
- CandiSift
CandiSift
What it is
CandiSift is an open-source applicant tracking system for parsing resumes, matching candidates to job descriptions, and ranking applicants through a bias-audit workflow. Recruiters can create roles, upload PDF, DOCX, or plain-text resumes, including multiple files, then review ranked results and accept or reject candidates. The workflow removes personally identifiable information before model processing, applies deterministic hard filters, ranks applicants, and sends those who remain to persona agents and human review. Token-cosine ranking, technical and risk persona agents, synthesis, and a recruiter interface support that process. Before confirming a staged batch, users can view an estimated processing cost; hard filters run before the paid LLM stage. CandiSift connects to Anthropic Claude through Agno persona agents, or uses an offline stub LLM without an API key. It can run in Docker, stores state in SQLite by default, and supports a Postgres database URL. The project describes security controls, an append-only audit log, and a bias-audit API endpoint. It has no shipped accuracy benchmark; OCR, embeddings/FAISS, and skillNER are documented swaps, not default capabilities.
For and against
For
- Accepts PDF, DOCX, and plain-text resumes
- Removes PII before model processing
- Shows estimated batch processing costs
- Keeps an append-only audit log
Against
- No shipped accuracy benchmark
- LLM processing can incur per-token costs
- OCR and embeddings are not default capabilities
The verdict
CandiSift combines resume parsing, filtering, ranking, and recruiter review in an open-source workflow. The absence of a shipped accuracy benchmark and possible LLM costs are important considerations.
Who it suits
CandiSift is intended for recruiters screening uploaded resumes, reviewing ranked candidates, and making accept-or-reject decisions. Teams should note that the project does not provide a shipped accuracy benchmark.
Plans and prices
Open-source self-hosted
No software fee stated; LLM processing can incur per-token costs
Read 4 Oct 2026 · github.com
On the record
- Bias audit
- The API includes a bias-audit endpoint covering pass-through and recommendation mix.github.com · read 4 Oct 2026
- Cost controls
- Users can see an estimated processing cost before confirming a staged batch, and deterministic hard filters run before the paid LLM stage.github.com · read 4 Oct 2026
- Deployment and storage
- The app can run in Docker, stores state in SQLite by default, and supports configuring a Postgres database URL.github.com · read 4 Oct 2026
- File formats
- Resume ingestion supports PDF, DOCX, and plain-text files, including multi-file uploads.github.com · read 4 Oct 2026
- Intended users
- The README describes a recruiter workflow for creating roles, screening uploaded resumes, reviewing ranked results, and accepting or rejecting candidates.github.com · read 4 Oct 2026
- LLM integration
- CandiSift integrates with Anthropic Claude through Agno persona agents; without an API key, it can use an offline stub LLM.github.com · read 4 Oct 2026
- Notable limits
- The project says it has no shipped accuracy benchmark and that OCR, embeddings/FAISS, and skillNER are documented swaps rather than included default capabilities.github.com · read 4 Oct 2026
- Privacy and audit
- CandiSift says it strips PII before a model sees candidate data and keeps an append-only audit log of scores, rationales, decisions, and versions.github.com · read 4 Oct 2026
- Product
- CandiSift is an open-source applicant tracking system for AI resume screening, CV parsing, job description and candidate matching, and bias-audited ranking.github.com · read 4 Oct 2026
- Ranking and review
- The product provides token-cosine ranking, technical and risk persona agents, synthesis, and a recruiter interface with ranked results and accept or reject decisions.github.com · read 4 Oct 2026
- Screening flow
- Its workflow parses resumes, strips personally identifiable information, applies hard filters, ranks candidates, and sends survivors to persona agents and human review.github.com · read 4 Oct 2026
- Security
- The repository describes HTTP Basic authentication, security headers, rate limiting, request body limits, upload validation, and prompt-injection defenses.github.com · read 4 Oct 2026
