ScoutLane — Recruitment Platform
AI-powered recruitment platform: public career pages, resume parsing, drag-and-drop pipelines, role-based admin, and the email/storage/webhook infrastructure a real hiring team needs.
Overview
ScoutLane is a recruitment platform that combines AI-powered resume parsing with full pipeline management: public career pages with custom application forms, a drag-and-drop Kanban admin dashboard, role-based access control, and the production infrastructure — async job workers, transactional email, cloud file storage, outbound webhooks — a real hiring team runs on day to day.
The Problem
Recruitment teams spend hours manually reviewing resumes and tracking candidates across spreadsheets. ScoutLane automates resume parsing and provides structured pipeline management to reduce time-to-hire, without leaving the operational gaps — auth, permissions, notifications, integrations — that turn a demo into unshippable software.
Questions Addressed
- 01
How can AI reliably extract structured data from unstructured resumes?
- 02
What pipeline views give recruiters the best visibility into candidate progress?
- 03
What does a resume-parsing pipeline need beyond the happy path to run unattended — retries, async workers, role separation?
Methodology
Resume Parsing Engine
Built an AI-powered resume parser (OpenRouter, model-configurable) that extracts education history, work experience, skills, and contact information from uploaded PDFs and DOCX files, normalized into structured candidate profiles and processed asynchronously via pg-boss workers so a slow parse never blocks the application flow.
Job Portal & Applications
Built public-facing career pages with department/location filters and custom application forms per job template. Candidates upload resumes which are automatically parsed and routed to the correct pipeline stage; Resend handles transactional email confirmations.
Admin Dashboard, RBAC & Integrations
Built a role-based admin dashboard (Admin / Recruiter / Hiring Manager via Auth.js JWT sessions) with drag-and-drop Kanban pipelines (dnd-kit), Recharts analytics, job template management, team management, and outbound webhook integrations for external systems — plus a full Vitest + Playwright test suite (236 tests, 36 files) and CI (lint → typecheck → test → build).
Key Results
Key Findings
AI resume parsing reduced manual data entry by ~80%, letting recruiters focus on candidate evaluation.
Kanban pipeline views improved team visibility into bottlenecks — stuck candidates became immediately visible.
Moving resume parsing to async pg-boss workers kept the application flow fast regardless of AI response latency — a lesson the earlier synchronous design didn't survive under real load.
Role-based access (Admin/Recruiter/Hiring Manager) turned out to matter more than any single feature — real hiring teams need permission boundaries before they'll trust a tool with candidate data.
Conclusion
ScoutLane bridges the gap between AI automation and human recruitment judgment. What started as a resume-parsing pipeline grew into full recruitment infrastructure — public career pages, RBAC, async workers, email, storage, webhooks, and a real test suite — because that's what shipping software for an actual hiring team requires, not just what a demo needs. Live at scoutlane.vercel.app.
Gallery
