Skip to content

PRODUCT.md

Latest commit

 

History

History
73 lines (53 loc) · 3.97 KB

PRODUCT.md

File metadata and controls

73 lines (53 loc) · 3.97 KB

Reqcore — Product Vision & Goals

The Sovereign Recruitment Engine

Reqcore is a lean, open-source Applicant Tracking System (ATS) designed to return power to the employer. We are the "Glass Box" alternative to the "Black Box" incumbents.

Problem Statement

Modern ATS platforms suffer from three structural problems:

  1. Data Hostage: Companies pay for access to their own candidate data. If the subscription lapses, the talent pool disappears.
  2. Opaque AI: Incumbent platforms use proprietary algorithms to rank candidates. Recruiters cannot see why a candidate was surfaced or rejected — creating legal and ethical liability.
  3. Per-Seat Tax: Adding a hiring manager or recruiter to the platform increases the software bill, punishing growing teams.

Unique Value Proposition (UVP)

1. Ownership over Access

You own the infrastructure (Postgres + MinIO). Your talent pool is a permanent asset — not a monthly subscription. Self-host on your own servers or use a managed deployment; either way, the data is yours.

2. Auditable Intelligence (The Glass Box)

We reject "Secret Algorithms." When AI ranks a candidate, it must provide a visible Matching Logic summary so recruiters can verify the result. Every AI decision is explainable and auditable.

3. The Anti-Growth Tax

No per-seat pricing. Reqcore is designed to let companies scale their hiring teams without increasing their software bill.

4. Privacy Sovereignty

By supporting local-first storage (MinIO) and local AI models (Ollama), Reqcore is the only ATS where sensitive candidate PII never has to leave the company's private network.

Target Users

Persona Description Primary Need
Recruiter Day-to-day user managing candidates and pipeline Fast candidate pipeline, transparent AI ranking
Hiring Manager Reviews candidates, makes hiring decisions Clear candidate comparisons, proof-based recommendations
HR Administrator Manages org settings, team access, compliance Multi-tenant control, data ownership, audit trails
Engineering/IT Deploys and maintains the system Simple self-hosting, Docker Compose, clear infra docs

Core Features (Current & Planned)

MVP — Foundation

  • Multi-tenant organizations (Better Auth + org plugin)
  • Job management (CRUD with status workflow: draft → open → closed → archived)
  • Candidate management (per-org candidate pool with deduplication by email)
  • Application tracking (link candidates to jobs, status workflow)
  • Document storage (resumes, cover letters via MinIO/S3)
  • Dashboard with pipeline overview
  • Organic SEO (sitemap, robots, JSON-LD structured data, blog content engine)

Phase 2 — Intelligence

  • Resume parsing (PDF → structured JSON)
  • AI candidate ranking with Matching Logic summary (Glass Box principle)
  • Skill extraction and matching
  • Local AI support via Ollama (privacy-first)

Phase 3 — Collaboration

  • Team comments and notes on candidates
  • Interview scheduling
  • Email integration (send/receive from within Reqcore)
  • Candidate portal (self-service application status)

Design Principles

  1. Show the Proof: If the AI matches a skill, highlight it. If a candidate is "High Potential," explain why based on the data.
  2. Recruiter-First UX: Every screen should answer "what do I need to do next?" within 3 seconds.
  3. Progressive Disclosure: Show summaries first, details on demand. Don't overwhelm with data.
  4. Tone: Professional, high-integrity, and engineering-grade. No marketing fluff in the UI.

Success Metrics

  • Time to first hire: How quickly can a new org go from setup to first candidate hired?
  • Transparency score: % of AI decisions with visible matching logic
  • Self-hosting success rate: % of deployments that complete without support tickets
  • Team adoption: Number of users per org (validates anti-seat-pricing model)