Step 2: Project Setup

# Create the project directory and navigate into it
mkdir job-application-agent
cd job-application-agent
 
# Create a virtual environment
python -m venv .venv
 
# Activate it (Linux/macOS)
source .venv/bin/activate
 
# Activate it (Windows)
# .venv\Scripts\activate

pip install langgraph langchain-openai python-dotenv

pip freeze > requirements.txt

# For Anthropic (Claude)
pip install langchain-anthropic
 
# For Google (Gemini)
pip install langchain-google-genai
 
# For local models (Ollama)
pip install langchain-ollama

# .env
OPENAI_API_KEY=sk-your-openai-key-here

# Create the directory structure
mkdir -p agents output
 
# Create the files
touch state.py graph.py utils.py main.py
touch agents/__init__.py agents/analyzer.py agents/tailor.py
touch agents/writer.py agents/reviewer.py

# utils.py
# ==========================================
# Shared utilities and configuration
# ==========================================
 
import os
from dotenv import load_dotenv
from langchain_openai import ChatOpenAI
 
load_dotenv()
 
def get_llm(model: str = "gpt-4o", temperature: float = 0.3) -> ChatOpenAI:
    """Create a configured LLM instance."""
    return ChatOpenAI(model=model, temperature=temperature)
 
 
def save_output(filename: str, content: str) -> str:
    """Save content to the output directory."""
    output_dir = os.path.join(os.path.dirname(__file__), "output")
    os.makedirs(output_dir, exist_ok=True)
 
    filepath = os.path.join(output_dir, filename)
    with open(filepath, "w", encoding="utf-8") as f:
        f.write(content)
 
    return filepath
 
 
def load_file(filepath: str) -> str:
    """Load a text file and return its contents."""
    with open(filepath, "r", encoding="utf-8") as f:
        return f.read()

# Alex Chen
 
**Software Engineer** | San Francisco, CA
alex.chen@email.com | github.com/alexchen
 
## Summary
 
Software engineer with 4 years of experience building web applications
and backend services. Focused on Python, JavaScript, and cloud infrastructure.
 
## Experience
 
### Software Engineer — Acme Corp (2022–Present)
- Built REST APIs serving 50k daily active users using Python and FastAPI
- Migrated legacy monolith to microservices architecture on AWS
- Led adoption of CI/CD pipelines, reducing deployment time by 60%
- Mentored 2 junior engineers through onboarding and code reviews
 
### Junior Developer — StartupXYZ (2020–2022)
- Developed React frontend for customer-facing dashboard
- Wrote integration tests that caught 30+ bugs before production
- Participated in on-call rotation and incident response
 
## Skills
 
Python, JavaScript, TypeScript, React, FastAPI, Django, PostgreSQL,
Redis, AWS (EC2, S3, Lambda), Docker, Git, CI/CD, REST APIs
 
## Education
 
B.S. Computer Science — UC Berkeley (2020)

# test_setup.py
# ==========================================
# Verify LangGraph and OpenAI are working
# ==========================================
 
from langchain_openai import ChatOpenAI
from dotenv import load_dotenv
 
load_dotenv()
 
llm = ChatOpenAI(model="gpt-4o", temperature=0)
response = llm.invoke("Say hello in exactly 5 words.")
 
print(response.content)
print("\nSetup verified — you're ready to build.")

python test_setup.py

rm test_setup.py

job-application-agent/
├── agents/
│   ├── __init__.py       ✅ Package init
│   ├── analyzer.py       📝 Empty
│   ├── tailor.py         📝 Empty
│   ├── writer.py         📝 Empty
│   └── reviewer.py       📝 Empty
├── output/               ✅ Output directory
├── state.py              📝 Empty
├── graph.py              📝 Empty
├── utils.py              ✅ Config and helpers
├── main.py               📝 Empty
├── sample_resume.md      ✅ Test resume
├── .env                  ✅ API key configured
└── requirements.txt      ✅ Dependencies saved