An Applicant Tracking System that uses Linear as its source of truth, with AI-powered candidate screening and job description generation.

• Framework: Next.js 16 with App Router
• Language: TypeScript
• Authentication: WorkOS AuthKit
• Data Source: Linear (Projects, Issues, Documents)
• AI Services: Cerebras (llama-3.3-70b)
• Storage: Upstash Redis
• Observability: Datadog (APM, Metrics, Logging)
• Testing: Vitest with fast-check for property-based testing
• UI: Radix UI + Tailwind CSS

• OAuth Integration: Connect your Linear workspace via OAuth 2.0
• Automated Job Publishing: Jobs automatically publish when Linear Projects are marked "In Progress"
• AI Job Descriptions: Generate structured job descriptions using Cerebras LLM based on your tone of voice
• Public Job Listings: Applicants can browse and apply to open positions
• AI Pre-screening: Automatic candidate evaluation comparing CVs against job requirements
• Document Parsing: Extract text from PDF and DOCX files for AI analysis
• Webhook Integration: Real-time sync with Linear via webhooks
• Subscription Management: Polar.sh integration for usage-based billing and subscription tiers
• Dynamic Pricing: Landing page pricing automatically loads from Polar API
• Full Observability: Comprehensive monitoring with Datadog APM, metrics, and logging

• Node.js 18+
• npm or yarn
• WorkOS account and API keys
• Linear workspace with OAuth app configured
• Cerebras API key
• Upstash Redis instance
• (Optional) Datadog account for observability

1. Clone the repository

2. Install dependencies:

   npm install

3. Copy the environment variables template:

   cp .env.example .env.local

4. Configure your environment variables in .env.local:

   • WorkOS: Authentication credentials
   • Linear: OAuth client ID, secret, and webhook secret
   • Cerebras: API key for AI inference
   • Upstash Redis: Connection URL and token
   • Polar: Access token, webhook secret, organization ID, and product IDs
   • Datadog: API key and service configuration (optional)

Run the development server:

npm run dev

Open http://localhost:3000 in your browser.

Run tests:

npm test

Run tests in watch mode:

npm run test:watch

Run tests with UI:

npm run test:ui

Build for production:

npm run build

Start production server:

npm start

├── app/                    # Next.js App Router pages
│   ├── api/               # API routes (auth, webhooks, health)
│   ├── dashboard/         # Authenticated dashboard
│   ├── jobs/              # Public job listings
│   └── onboarding/        # Linear integration setup
├── components/            # React components
│   ├── ui/               # Radix UI components
│   ├── application-form.tsx
│   ├── initiative-selector.tsx
│   └── tone-of-voice-setup.tsx
├── lib/                   # Core business logic
│   ├── actions/          # Server actions
│   ├── cerebras/         # AI inference (job descriptions, screening)
│   ├── datadog/          # Observability (APM, metrics, logging)
│   ├── linear/           # Linear SDK integration
│   └── utils/            # Utility functions (retry logic, etc.)
├── types/                 # TypeScript type definitions
└── middleware.ts          # Auth and correlation ID middleware

The system uses Linear as the single source of truth:

• Initiatives: Hiring containers (ATS Container)
• Projects: Job openings (published when status is "In Progress")
• Issues: Candidates/applicants (with workflow states: Triage → Screening/Declined)
• Documents: CVs, cover letters, tone of voice guides

1. Onboarding: User authenticates via WorkOS → connects Linear via OAuth → selects/creates ATS Container Initiative → sets up tone of voice
2. Job Publishing: Project marked "In Progress" → AI generates job description → published to public site
3. Application: Applicant submits form → CV parsed → Linear Issue created → AI pre-screening triggered → Issue state updated based on match quality
4. Sync: Linear webhooks keep job listings and candidate states in sync

• Job Description Generation: Uses Cerebras llama-3.3-70b to create structured job descriptions based on project details and organizational tone of voice
• Candidate Pre-screening: Compares CV content against job requirements to automatically triage candidates into Screening, Triage, or Declined states

Datadog integration provides:

• APM Tracing: Distributed traces across all API calls and external services
• Custom Metrics: LLM latency, token usage, webhook processing times
• Error Logging: Full context with stack traces and correlation IDs
• Alerting: Critical operation failures trigger Datadog events

ISC