Last updated: January 6

Let's Build: AI resume parser with Transloadit

Kevin van Zonneveld

Co-founder · Amsterdam, The Netherlands ·

Resume parsing is a classic SaaS problem: take messy, real-world documents and extract structured fields that feed ATS pipelines. With Transloadit, you can build that workflow from primitives in a single assembly.

What we’re building

Accept PDF, DOCX, or image resumes
Convert non-PDFs to PDF with 🤖 /document/convert
Extract fields with 🤖 /ai/chat

Schema first (zod)

import { z } from 'zod'
import { zodToJsonSchema } from 'zod-to-json-schema'

const resumeSchema = z.object({
  full_name: z.string().optional(),
  headline: z.string().optional(),
  email: z.string().optional(),
  phone: z.string().optional(),
  location: z.string().optional(),
  summary: z.string().optional(),
  skills: z.array(z.string()).optional(),
  work_experience: z
    .array(
      z.object({
        company: z.string().optional(),
        title: z.string().optional(),
        start_date: z.string().optional(),
        end_date: z.string().optional(),
        highlights: z.array(z.string()).optional(),
      }),
    )
    .optional(),
  education: z
    .array(
      z.object({
        institution: z.string().optional(),
        degree: z.string().optional(),
        start_date: z.string().optional(),
        end_date: z.string().optional(),
      }),
    )
    .optional(),
})

const toJsonSchema = (schema: z.ZodTypeAny): unknown =>
  typeof (z as { toJSONSchema?: typeof zodToJsonSchema }).toJSONSchema === 'function'
    ? (z as { toJSONSchema: typeof zodToJsonSchema }).toJSONSchema(schema)
    : zodToJsonSchema(schema)

const resumeJsonSchema = toJsonSchema(resumeSchema)

Pipeline overview

PDF resumes ──▶ /ai/chat
Other docs  ──▶ /document/convert (pdf) ──▶ /ai/chat

Assembly snippet

const assembly = await transloadit.createAssembly({
  params: {
    steps: {
      pdf_verified: {
        robot: '/file/filter',
        use: ':original',
        accepts: [['${file.mime}', '==', 'application/pdf']],
      },
      non_pdf: {
        robot: '/file/filter',
        use: ':original',
        accepts: [['${file.mime}', '!=', 'application/pdf']],
      },
      pdf_converted: {
        robot: '/document/convert',
        use: 'non_pdf',
        format: 'pdf',
      },
      extract_resume: {
        robot: '/ai/chat',
        use: ['pdf_verified', 'pdf_converted'],
        model: 'anthropic/claude-4-sonnet-20250514',
        format: 'json',
        schema: JSON.stringify(resumeJsonSchema),
        messages: 'Extract resume fields. Omit keys if unknown.',
        result: true,
      },
    },
  },
  files: {
    document: filePath,
  },
  waitForCompletion: true,
})

Try the example app

The runnable version of this post lives at example_apps/resume-parser/run.ts.

node example_apps/resume-parser/run.ts

Suggested inputs for visuals

Use the bundled samples or swap in a real resume PDF:

_assets/demos/inputs/first_document.pdf
_assets/demos/inputs/second_document.pdf
_assets/demos/inputs/scan.pdf

Next steps

Add a normalization layer for phone numbers and dates.
Flag missing fields and route to manual review.
Store results in a talent database or ATS.