Skip to main content

​
System Requirements

Before installing, ensure your system meets these requirements:

Python Version

Python 3.10 or 3.11 (recommended)

RAM

Minimum 8GB RAM (16GB recommended for large candidate pools)

Storage

~2GB free space for dependencies and embedding models

Operating System

Windows, macOS, or Linux
Python 3.12+ is not yet officially supported by all dependencies. Stick with Python 3.10 or 3.11 for best compatibility.

​
Installation Methods

​
Dependencies Breakdown

Here’s what gets installed from requirements.txt: 
langchain
langchain-community
langchain-google-genai
langchain-huggingface
sentence-transformers
faiss-cpu
pypdf
reportlab
pandas==2.2.2
matplotlib
plotly

​
Core Dependencies

PackageVersionPurpose
langchainlatestFramework for building LLM applications
langchain-google-genailatestGemini 1.5 Flash integration
langchain-huggingfacelatestHuggingFace embeddings wrapper
sentence-transformerslatestPre-trained embedding models
faiss-cpulatestVector similarity search engine
pypdflatestPDF document parsing
pandas2.2.2Structured data manipulation
GPU Acceleration: If you have a CUDA-compatible GPU, replace faiss-cpu with faiss-gpu for faster vector search on large candidate databases.

​
API Key Setup

The system requires a Google API key to use Gemini 1.5 Flash. Follow these steps to obtain and configure it:

​
Step 1: Get Your API Key

1

Visit Google AI Studio

Navigate to Google AI Studio
2

Create API Key

Click “Create API Key” and select a Google Cloud project (or create a new one)
3

Copy the Key

Copy the generated API key - you’ll use it in the next step
Google AI Studio offers a generous free tier with 60 queries per minute for Gemini 1.5 Flash. Perfect for testing and small-scale deployments.

​
Step 2: Configure Environment Variable

Set the GOOGLE_API_KEY environment variable in your system: 
# Temporary (current session only)
export GOOGLE_API_KEY="your_api_key_here"

# Permanent (add to ~/.bashrc or ~/.zshrc)
echo 'export GOOGLE_API_KEY="your_api_key_here"' >> ~/.bashrc
source ~/.bashrc
Security Best Practices:
  • Never commit .env files to version control (add to .gitignore)
  • Use different API keys for development and production
  • Rotate keys periodically
  • Monitor API usage in Google Cloud Console

​
Verification Steps

After installation, verify everything is working correctly:

​
1. Check Python Version

python --version
# Expected: Python 3.10.x or 3.11.x

​
2. Verify Dependencies

pip list | grep -E "langchain|faiss|sentence-transformers"
Expected Output: 
faiss-cpu                 1.7.4
langchain                 0.1.0
langchain-community       0.0.13
langchain-google-genai    0.0.5
langchain-huggingface     0.0.1
sentence-transformers     2.2.2

​
3. Test API Connection

Create a test script test_setup.py:
test_setup.py
import os
from langchain_google_genai import ChatGoogleGenerativeAI
from langchain_huggingface import HuggingFaceEmbeddings

# Test 1: API Key
print("[1/3] Checking API key...")
if not os.getenv("GOOGLE_API_KEY"):
    raise ValueError("❌ GOOGLE_API_KEY not found in environment variables")
print("✓ API key configured")

# Test 2: LLM Connection
print("\n[2/3] Testing Gemini connection...")
try:
    llm = ChatGoogleGenerativeAI(
        model="gemini-1.5-flash",
        temperature=0
    )
    response = llm.invoke("Hello, respond with just 'OK' if you can read this")
    print(f"✓ Gemini response: {response.content}")
except Exception as e:
    print(f"❌ Gemini connection failed: {e}")
    raise

# Test 3: Embeddings
print("\n[3/3] Testing embeddings model...")
try:
    embeddings = HuggingFaceEmbeddings()
    test_vec = embeddings.embed_query("test sentence")
    print(f"✓ Embeddings working (dimension: {len(test_vec)})")
except Exception as e:
    print(f"❌ Embeddings failed: {e}")
    raise

print("\n✅ All systems operational!")
Run the test: 
python test_setup.py
Expected Output: 
[1/3] Checking API key...
✓ API key configured

[2/3] Testing Gemini connection...
✓ Gemini response: OK

[3/3] Testing embeddings model...
✓ Embeddings working (dimension: 768)

✅ All systems operational!

​
4. Verify FAISS Installation

import faiss
import numpy as np

# Create a simple vector index
dimension = 128
index = faiss.IndexFlatL2(dimension)

# Add some random vectors
vectors = np.random.random((10, dimension)).astype('float32')
index.add(vectors)

print(f"✓ FAISS working - indexed {index.ntotal} vectors")

​
Troubleshooting

Solution: Install the CPU version explicitly:
pip uninstall faiss faiss-cpu faiss-gpu
pip install faiss-cpu
If you have a GPU:
pip install faiss-gpu
Solution: Update certificates or disable SSL verification (not recommended for production):
pip install --upgrade certifi
Or set environment variable:
export CURL_CA_BUNDLE=""
Solution: Ensure the key is properly set and restart your terminal:
import os
print(os.getenv("GOOGLE_API_KEY"))  # Should print your key
If it returns None, the variable isn’t set. Re-run the export command and restart your Python session.
Solution: Reduce batch size or use a smaller embedding model:
embeddings = HuggingFaceEmbeddings(
    model_name="sentence-transformers/all-MiniLM-L6-v2"  # Smaller model
)
Solution: The project specifies pandas==2.2.2 for stability. If you encounter issues:
pip install --upgrade pandas==2.2.2

​
Optional: GPU Acceleration

For production deployments with large candidate databases (1000+ resumes), GPU acceleration significantly improves performance:
1

Install CUDA Toolkit

Download from NVIDIA CUDA Downloads
2

Install GPU Version of FAISS

pip uninstall faiss-cpu
pip install faiss-gpu
3

Verify GPU Detection

import faiss
print(f"GPU available: {faiss.get_num_gpus()}")
GPU acceleration provides 10-100x speedup for vector search on large datasets. However, the free Google Colab environment already includes GPU support, making it unnecessary for most users.

​
Next Steps

You’re all set! Here’s what to do next:

Run the Quickstart

Try the system with sample data in 5 minutes

Architecture Guide

Understand how components work together

Configuration

Customize models, prompts, and retrieval settings

API Reference

Explore available functions and classes
Installation Complete! You now have a working local environment for the RAG Recruitment Assistant.

Build docs developers (and LLMs) love

Get started for freeTalk to us