the files

agents.py

@@ -0,0 +1,45 @@
+from crewai import Agent
+
+def create_resume_analyst(llm):
+    return Agent(
+        llm=llm,
+        role="Resume Analyst",
+        goal="Analyze resumes and provide detailed feedback",
+        backstory="You're an experienced HR professional specialized in resume screening "
+                  "and providing constructive feedback to candidates.",
+        allow_delegation=False,
+        verbose=True
+    )
+
+def create_feedback_specialist(llm):
+    return Agent(
+        llm=llm,
+        role="Feedback Specialist",
+        goal="Provide actionable improvement suggestions",
+        backstory="You're an expert in career development and resume optimization, "
+                  "focusing on providing specific, actionable feedback.",
+        allow_delegation=False,
+        verbose=True
+    )
+
+def create_hr_problem_analyst(llm):
+    return Agent(
+        llm=llm,
+        role="HR Problem Analyst",
+        goal="Analyze HR challenges and provide practical solutions",
+        backstory="You're an experienced HR consultant specialized in providing "
+                  "actionable advice for HR-related questions and challenges.",
+        allow_delegation=False,
+        verbose=True
+    )
+
+def create_solution_architect(llm):
+    return Agent(
+        llm=llm,
+        role="Solution Architect",
+        goal="Provide detailed, implementable solutions",
+        backstory="You're an HR solution expert who provides specific, "
+                  "practical solutions with examples and best practices.",
+        allow_delegation=False,
+        verbose=True
+    )

tasks.py

@@ -0,0 +1,70 @@
+from crewai import Task, Crew
+from agents import (
+    create_resume_analyst,
+    create_feedback_specialist,
+    create_hr_problem_analyst,
+    create_solution_architect
+)
+
+def create_resume_review_crew(resume_text, llm):
+    resume_analyst = create_resume_analyst(llm)
+    feedback_specialist = create_feedback_specialist(llm)
+
+    analyze_resume = Task(
+        description=(
+            f"Analyze this resume:\n{resume_text}\n"
+            "1. Evaluate the overall structure and format\n"
+            "2. Assess the content quality and relevance\n"
+            "3. Identify strengths and weaknesses\n"
+            "4. Check for essential components"
+        ),
+        expected_output="Detailed resume analysis",
+        agent=resume_analyst
+    )
+    provide_feedback = Task(
+        description=(
+            "Based on the analysis, provide detailed feedback:\n"
+            "1. List specific improvements needed\n"
+            "2. Highlight positive aspects\n"
+            "3. Suggest concrete changes\n"
+            "4. Provide formatting recommendations"
+        ),
+        expected_output="Comprehensive feedback with actionable suggestions",
+        agent=feedback_specialist
+    )
+    return Crew(
+        agents=[resume_analyst, feedback_specialist],
+        tasks=[analyze_resume, provide_feedback],
+        verbose=2
+    )
+
+def create_hr_crew(question, llm):
+    hr_analyst = create_hr_problem_analyst(llm)
+    solution_architect = create_solution_architect(llm)
+
+    analyze_question = Task(
+        description=(
+            f"Analyze this HR question or challenge: {question}\n"
+            "1. Identify the core issue\n"
+            "2. Consider relevant HR best practices\n"
+            "3. Note any potential complications"
+        ),
+        expected_output="Clear analysis with key considerations",
+        agent=hr_analyst
+    )
+    provide_solution = Task(
+        description=(
+            f"Based on the analysis, address this question: {question}\n"
+            "1. Provide specific, actionable solutions\n"
+            "2. Include relevant examples\n"
+            "3. Add implementation tips\n"
+            "4. Consider different organizational contexts"
+        ),
+        expected_output="Detailed, practical solution with examples",
+        agent=solution_architect
+    )
+    return Crew(
+        agents=[hr_analyst, solution_architect],
+        tasks=[analyze_question, provide_solution],
+        verbose=2
+    )