Understanding AI job analysis

What AI analysis does

When you import a job, our AI:

1. Reads the entire job posting or document
2. Identifies key information fields
3. Extracts relevant data into structured fields
4. Organizes everything for easy review and editing

This happens in 10-30 seconds, saving you from manual data entry.

Fields we extract

Job title and level

The AI identifies:

• Job title (e.g., "Backend Engineer", "Product Designer")
• Seniority level (Junior, Mid, Senior, Staff, Principal, Lead)
• Team or department context

Common extractions:

• "Senior Software Engineer" → Title: Software Engineer, Level: Senior
• "Staff Backend Developer" → Title: Backend Developer, Level: Staff

Tech stack

Technologies, languages, and tools mentioned:

• Programming languages (Python, TypeScript, Go, etc.)
• Frameworks (React, Django, Rails, etc.)
• Infrastructure (AWS, Kubernetes, Docker, etc.)
• Databases (PostgreSQL, MongoDB, Redis, etc.)

The AI captures both required and nice-to-have technologies.

Location and remote policy

Work arrangement details:

• Office locations (city, country)
• Remote status (fully remote, hybrid, on-site)
• Timezone requirements
• Relocation support

Compensation

Salary and benefits information:

• Base salary range
• Equity or stock options
• Bonus structure
• Benefits highlights

Note: Many job postings omit compensation. If not found, you will need to add it manually. We strongly recommend including compensation as it significantly increases developer engagement.

Experience requirements

What the role requires:

• Years of experience
• Must-have skills
• Educational requirements (if mentioned)
• Specific domain experience

Job description

The full scope of the role:

• Responsibilities
• Team context
• Company information
• Growth opportunities

Confidence indicators

After extraction, you may see indicators showing AI confidence:

• High confidence: Field is clearly stated in the source
• Medium confidence: Field is inferred from context
• Low confidence / flagged: AI is uncertain, please verify

Always review flagged fields carefully.

What AI cannot do

The AI extracts what is in the source. It cannot:

• Add compensation if not mentioned
• Determine remote policy if not stated
• Know internal details not in the posting
• Read behind paywalls or logins

If information is missing, add it manually in the editor.

Improving extraction quality

For best results, ensure your source includes:

• Clear job title and level
• Explicit tech stack requirements
• Specific location or remote policy
• Compensation range (highly recommended)
• Detailed responsibilities

The more complete your source, the better the extraction.

Verifying results

After AI analysis:

1. Read the preview: Does it look accurate?
2. Check critical fields: Title, stack, location, compensation
3. Fix any errors: Edit directly in the form
4. Add missing info: Complete any empty required fields

The side-by-side editor shows your edits in real-time in the preview.

Troubleshooting

AI extracted wrong information:

• Edit the field directly in the form
• The AI learns from common job posting patterns but is not perfect
• Unusual formatting may cause extraction errors

Important fields are empty:

• The source likely did not include that information
• Add the details manually
• Consider updating your original job posting for future imports