InterviewAgent

Multi-agent pipeline that automates the end-to-end job application workflow — resume tailoring, cover letter generation, and application submission at scale.

What It Does

InterviewAgent runs a coordinated pipeline of AI agents to handle the tedious parts of job searching:

1. Job Discovery — scrapes and filters job postings matching your criteria
2. Resume Optimizer — rewrites your resume to match each job's keywords and requirements
3. Cover Letter Generator — writes targeted cover letters using Claude with company/role context
4. Application Submitter — fills and submits applications via Playwright browser automation
5. Tracking & Reporting — logs every application to Supabase with status, match score, and timestamps

At full throughput: 500+ applications/day.

Tech Stack

Layer	Tech
AI / LLM	Claude (Anthropic), MCP tools
Orchestration	Claude Code, multi-agent pipeline
Browser Automation	Playwright
Storage	Supabase (PostgreSQL)
Language	Python

Architecture

Job Sources → Discovery Agent
                    ↓
              Job Queue (Supabase)
                    ↓
         ┌──────────┴──────────┐
   Resume Agent         Cover Letter Agent
         └──────────┬──────────┘
                    ↓
           Submission Agent (Playwright)
                    ↓
           Tracking DB (Supabase)

Quick Start

# Clone
git clone https://github.com/hansraj316/InterviewAgent
cd InterviewAgent

# Install dependencies
pip install -r requirements.txt
playwright install chromium

# Configure environment
cp .env.example .env
# Set: ANTHROPIC_API_KEY, SUPABASE_URL, SUPABASE_ANON_KEY

# Run
python main.py

Configuration

Edit config.yaml to set:

• Job search keywords and location
• Target companies/industries
• Daily application limit
• Resume base file path
• Cover letter tone preferences

Indian Avengers Context

InterviewAgent is part of the Indian Avengers autonomous AI org. It's operated by the InterviewAgent sub-agent and monitored via Mission Control.

Agent activity is logged to ~/.openclaw/workspace-work/InterviewAgent/autonomy-log.jsonl.

License

MIT

Daily TPM delivery update (2026-04-22)

• Functional: Add adaptive interview flows that adjust question depth by candidate responses
• Non-functional: Expand test coverage for scoring rubric consistency across roles

Daily delivery update (2026-04-28)

• Functional: Add role-specific interview kit generator with rubric presets
• Non-functional: Improve transcript parsing accuracy for noisy audio inputs