Skip to main content

Documentation Index

Fetch the complete documentation index at: https://mintlify.com/vrashmanyu605-eng/Langchain_Interview_Multi_Agents_Flow/llms.txt

Use this file to discover all available pages before exploring further.

The matching agent is the first agent that requires outputs from both parsing agents. It takes the structured candidate_profile and jd_analysis and asks the LLM to evaluate the candidate across four dimensions, then produce a single fit score. The resulting matching_analysis is a key input for the evaluation agent, which uses it alongside interview question banks to form the final hiring recommendation.

​
Source code

from llm import llm


def matching_agent(state):

    candidate_profile = state["candidate_profile"]

    jd_analysis = state["jd_analysis"]

    response = llm.invoke(
        f"""
        You are a Candidate Matching Agent.

        Compare the candidate profile with the job description.

        Candidate Profile:
        {candidate_profile}

        Job Description Analysis:
        {jd_analysis}

        Evaluate:
        - technical fit
        - experience fit
        - leadership fit
        - communication fit

        Return:
        - fit_score
        - strengths
        - weaknesses
        - missing_skills
        - recommendation

        Return structured JSON only.
        """
    )

    return {
        "matching_analysis": response.content
    }

​
Inputs

​
candidate_profile
string
required
The structured JSON profile produced by the resume parser agent.
​
jd_analysis
string
required
The structured JSON analysis produced by the JD analysis agent.

​
Output

​
matching_analysis
string
A JSON string containing the fit score, strengths, weaknesses, missing skills, and a routing recommendation. Read by the evaluation agent.

​
Four evaluation dimensions

The prompt asks the LLM to evaluate the candidate across four dimensions before producing the overall score:
DimensionWhat it assesses
Technical fitOverlap between the candidate’s skills and frameworks versus the JD’s required and preferred skills
Experience fitWhether the candidate’s years of experience and seniority level match the role’s expectations
Leadership fitWhether the candidate’s background meets any people management or mentoring requirements in the JD
Communication fitInferred from the resume’s project descriptions, roles, and context — does the candidate appear to operate at the right communication level for the role

​
Example output

{
  "fit_score": 78,
  "strengths": ["Strong Python skills", "Relevant cloud experience"],
  "weaknesses": ["No Kubernetes experience"],
  "missing_skills": ["Kubernetes", "GraphQL"],
  "recommendation": "Proceed to technical interview with focus on system design"
}

​
State diagram

Requires: candidate_profile, jd_analysis
    ↓
matching_agent
    ↓
Provides: matching_analysis

Build docs developers (and LLMs) love

Get started for freeTalk to us