AI Resume Screening Platform
An in-development applicant tracking system that helps recruiters manage job openings, candidates, and interview pipelines in one place — with AI-assisted resume analysis layered in as it matures. The core ATS workflows are being built first; AI capabilities are added carefully and only marked done once they actually work.
The problem
Reviewing large volumes of resumes by hand is slow and inconsistent. Recruiters need a structured way to track candidates through a pipeline, with optional AI assistance to surface relevant profiles faster.
Who it's for
- Recruiters and hiring teams
- Small companies running their own hiring
User roles
- Recruiter
My responsibility
What I actually did on this project — kept honest, especially on collaborative work.
- Designing the ATS data model and recruiter workflows
- Building job, candidate, and pipeline management
- Planning the AI-assisted analysis layer in an honest, incremental way
Product decisions
- Build reliable ATS fundamentals before any AI features
- Treat AI as an assistive layer, not a black-box gatekeeper
Architecture
Next.js App Router frontend with a Firebase/Firestore backend for jobs, candidates, and pipeline state. The AI-analysis layer is designed as a separate, optional service so the ATS works fully without it.
Key features
Every feature is labelled by its real state. Nothing planned is shown as shipped.
- Job opening managementImplemented
- Candidate profilesImplemented
- Resume uploadImplemented
- Candidate filteringPrototype
- Candidate status pipelinesPrototype
- Interview trackingPrototype
- Dashboard analyticsPrototype
- AI-assisted resume analysisPlanned
- Job-to-candidate matchingPlanned
- Recruiter workflow automationPlanned
Challenges
- Keeping AI claims honest while the analysis layer is still being built
- Designing pipelines flexible enough for different hiring processes
Solutions
- Explicit per-feature states so nothing AI-related is shown as done prematurely
- A configurable status pipeline rather than a fixed flow
Current limitations
What this project does not do yet — stated plainly.
- AI-assisted analysis and matching are planned, not yet implemented
- Filtering, pipelines, and analytics are at prototype stage
Lessons learned
- Shipping the non-AI core first makes the AI layer easier to evaluate honestly
Future roadmap
- Implement AI-assisted resume analysis
- Add job-to-candidate matching
- Automate recruiter workflows