Task Decomposition

The problem with single-shot extraction

Asking a single LLM call to populate the entire JSON Resume schema in one shot has several problems:

• High cognitive load — the model must simultaneously extract contact info, work history, education, skills, languages, and projects from a long document
• Schema echo — small models generate a syntactically perfect but empty structure
• Cross-section pollution — projects end up in work experience, languages end up in skills
• Token waste — the full CV text is sent once, but most of it is irrelevant to each section

How task decomposition works

resume-intel runs parallel focused extractions per section. By default, 8 sections run concurrently:

Resume text
    ↓
┌─────────┬──────┬───────────┬────────┬───────────┬──────────┬──────────────┬───────────┐
│ basics  │ work │ education │ skills │ languages │ projects │ certificates │ volunteer │
│ prompt  │ ...  │ ...       │ ...    │ ...       │ ...      │ ...          │ ...       │
│ schema  │ ...  │ ...       │ ...    │ ...       │ ...      │ ...          │ ...       │
│maxTokens│ ...  │ ...       │ ...    │ ...       │ ...      │ ...          │ ...       │
└─────────┴──────┴───────────┴────────┴───────────┴──────────┴──────────────┴───────────┘
    ↓         ↓        ↓         ↓          ↓          ↓             ↓             ↓
  result    result   result    result     result     result         result        result
    ↓
  merge → deduplicate → validate → return

Each section runs concurrently. A failure in one section does not block the others.

Choosing which sections to extract

By default, 8 sections are extracted. You can override this with the sections option:

// Extract only what you need — saves tokens and time
const result = await parseResume(buffer, {
  model,
  sections: ['basics', 'work', 'education'],
})
 
// Extract all 12 available sections
import { ALL_SECTIONS } from '@edwinfom/resume-intel'
const result = await parseResume(buffer, {
  model,
  sections: [...ALL_SECTIONS],
})

All 12 available sections

Per-section retry

If a section fails Zod validation, it retries independently with the specific error fed back to the LLM:

Section extraction attempt 1
    fails validation: "work.0.startDate must be YYYY-MM format"
    ↓
Section extraction attempt 2 (with correction prompt)
    passes
    ↓
Result merged

A bad languages extraction does not cause work to retry — only languages retries.

Observability

const result = await parseResume(pdfBuffer, { model })
 
console.log(result.meta.sectionsRequested)
// ['basics', 'work', 'education', 'skills', 'languages', 'projects', 'certificates', 'volunteer']
 
for (const s of result.meta.sectionResults ?? []) {
  const status  = s.success ? 'OK  ' : 'FAIL'
  const retries = s.retryCount > 0 ? ` (${s.retryCount} retries)` : ''
  console.log(`${status} ${s.section}${retries}`)
}

Disabling task decomposition

For simple single-column CVs or when using a custom outputSchema, single-shot extraction is faster:

const result = await parseResume(pdfBuffer, {
  model,
  useTaskDecomposition: false,
})

Configuring retries

const result = await parseResume(pdfBuffer, {
  model,
  maxRetries: 2, // per-section retry limit (default: 3)
})