Job_application.py

# Import required libraries
import os
import streamlit as st
from crewai import Agent, Task, Crew, LLM
from crewai_tools import (
    SerperDevTool,
    FileReadTool,
    MDXSearchTool,
    ScrapeWebsiteTool
)
from embedchain import App
from embedchain.embedder import GeminiEmbedder

# Set up API keys
gemini_api_key = "AIzaSyDzsGJCnE3017fYnOM0Fp_aWs4YtN4d4TI"
serper_api_key = "b86545fdabc35dcb13fd8cc0a9b88c3a17b6dc89"

# Configure environment variables
os.environ["GEMINIAI_API_KEY"] = gemini_api_key
os.environ["SERPER_API_KEY"] = serper_api_key

# Initialize Gemini embedding model
gemini_embedder = GeminiEmbedder(api_key=gemini_api_key)

# Initialize EmbedChain App with Gemini
embedchain_app = App(embedding_model=gemini_embedder)

# Initialize Tools
search_tool = SerperDevTool()
scrape_tool = ScrapeWebsiteTool()
resume_file_path = 'resume.md'  # Use a relative or dynamic file path
read_resume = FileReadTool(file_path=resume_file_path)
semantic_search_resume = MDXSearchTool(mdx=resume_file_path, embedding_model=gemini_embedder, app=embedchain_app)

# Agent 1: Researcher
researcher = Agent(
    role="Tech Job Researcher",
    goal="Analyze job postings and extract required qualifications.",
    tools=[scrape_tool, search_tool],
    verbose=True,
    backstory=(
        "An expert in analyzing job postings, you identify essential skills "
        "and qualifications required for job applications."
    )
)

# Agent 2: Profiler
profiler = Agent(
    role="Personal Profiler for Engineers",
    goal="Create a detailed profile for job applicants.",
    tools=[read_resume, semantic_search_resume],
    verbose=True,
    backstory=(
        "Specializing in building comprehensive profiles, you extract and "
        "synthesize information to create impactful resumes."
    )
)

# Agent 3: Resume Strategist
resume_strategist = Agent(
    role="Resume Strategist for Engineers",
    goal="Refine resumes to align with job requirements.",
    tools=[read_resume, semantic_search_resume],
    verbose=True,
    backstory=(
        "Your expertise lies in crafting resumes that highlight key skills "
        "and experiences to match job requirements."
    )
)

# Agent 4: Interview Preparer
interview_preparer = Agent(
    role="Interview Preparer",
    goal="Generate potential interview questions and talking points.",
    tools=[read_resume, semantic_search_resume],
    verbose=True,
    backstory=(
        "You prepare candidates for interviews by formulating relevant questions "
        "and talking points based on the job and their profile."
    )
)

# Define Tasks
research_task = Task(
    description="Analyze the job posting URL to extract key skills, qualifications, and requirements.",
    expected_output="A structured list of job requirements.",
    agent=researcher,
    async_execution=True
)

profile_task = Task(
    description="Create a detailed profile from the resume and personal write-up.",
    expected_output="A comprehensive profile document.",
    agent=profiler,
    async_execution=True
)

resume_strategy_task = Task(
    description="Tailor the resume based on job requirements and personal profile.",
    expected_output="An updated resume tailored to the job.",
    output_file="tailored_resume.md",
    context=[research_task, profile_task],
    agent=resume_strategist
)

interview_preparation_task = Task(
    description="Generate interview questions and talking points based on the tailored resume.",
    expected_output="A document with key interview questions and talking points.",
    output_file="interview_materials.md",
    context=[research_task, profile_task, resume_strategy_task],
    agent=interview_preparer
)

# Crew Setup
job_application_crew = Crew(
    agents=[researcher, profiler, resume_strategist, interview_preparer],
    tasks=[research_task, profile_task, resume_strategy_task, interview_preparation_task],
    verbose=True
)

# Streamlit Application
st.title("AI-Powered Job Application Assistant")

# User Inputs
st.header("Provide Job Details")
job_posting_url = st.text_input("Job Posting URL", "https://jobs.lever.co/AIFund/6c82e23e-d954-4dd8-a734-c0c2c5ee00f1")
github_url = st.text_input("GitHub Profile URL", "https://github.com/joaomdmoura")
personal_writeup = st.text_area(
    "Personal Writeup",
    """Noah is an accomplished Software Engineering Leader with 18 years of experience, 
    specializing in managing remote and in-office teams. He holds an MBA and has a strong 
    background in AI and data science. Noah has successfully led major tech initiatives 
    and startups, driving innovation and growth."""
)

# File Upload for Resume
st.header("Upload Resume")
uploaded_resume = st.file_uploader("Upload your resume (Markdown format)", type=["md"])
if uploaded_resume:
    resume_file_path = os.path.join("uploads", uploaded_resume.name)
    with open(resume_file_path, "wb") as f:
        f.write(uploaded_resume.getbuffer())
    st.success("Resume uploaded successfully!")

# Start Job Application Process
if st.button("Start Job Application Process"):
    job_application_inputs = {
        'job_posting_url': job_posting_url,
        'github_url': github_url,
        'personal_writeup': personal_writeup
    }

    # Run the Crew
    result = job_application_crew.kickoff(inputs=job_application_inputs)
    st.success("Job Application Process Completed!")

    # Display Results
    if os.path.exists("tailored_resume.md"):
        st.header("Generated Tailored Resume")
        with open("tailored_resume.md", "r") as f:
            st.markdown(f.read(), unsafe_allow_html=True)

    if os.path.exists("interview_materials.md"):
        st.header("Generated Interview Materials")
        with open("interview_materials.md", "r") as f:
            st.markdown(f.read(), unsafe_allow_html=True)