Job Analysis

Step 3: Extract Requirements

Categorize every requirement into one of these groups:

Key Responsibilities

List the main duties and responsibilities described in the role. For each, note:

• The core activity
• Domain or context (e.g., "cloud infrastructure", "customer-facing")
• Scale indicators (team size, budget, user count)

Required Qualifications (Must-Have)

Skills and qualifications explicitly marked as required. For each:

• Skill or qualification name
• Years of experience if specified
• Context (e.g., "Python for data pipelines" vs. just "Python")

Preferred Qualifications (Nice-to-Have)

Skills marked as "preferred", "bonus", "nice to have", or "a plus". Same format as above.

ATS Keywords

Extract all technical terms, tools, methodologies, certifications, and domain-specific phrases that an ATS system would likely scan for. Include:

• Exact phrases from the posting (preserve original wording)
• Acronyms and their expansions
• Tool/platform names with version numbers if mentioned

Step 4: Identify Themes

Identify 3–5 overarching themes the role emphasizes. Examples:

• "Cross-functional leadership"
• "Data-driven decision making"
• "Cloud-native architecture"
• "Developer experience and tooling"
• "Stakeholder communication"

Step 5: Priority Rating

Rate each extracted requirement:

• Critical — Explicitly required, mentioned multiple times, or in the job title
• Important — Listed in requirements, mentioned once
• Nice-to-have — Listed as preferred or bonus

Step 6: Output

Return the structured analysis using the categories above. Include the detected language of the job description prominently in the metadata block — this is used by the orchestrator to determine whether to prompt for CV output language. End with a "Tailoring Recommendations" section: 3–5 bullet points advising which aspects of a candidate's profile should be emphasized most.

Constraints

• Do NOT add requirements that are not in the job description.
• Do NOT interpret vague language as specific requirements (e.g., "team player" stays as-is, don't expand it into specific skills).
• Preserve the original language of the job posting when extracting keywords.
• If the job description is very short or vague, note this as a limitation.