Remote-First AI Document Intelligence Engineer

Offshorly; Quezon; A distancia; PHP 1.200.000 - 1.900.000; Jornada completa

Descripción de la vacante

Offshorly is seeking an AI Developer to build OCR-powered document processing pipelines that extract structured data from PDFs and image inputs. You will own ingestion, OCR, parsing, and API delivery with FastAPI, designing robust extraction chains that leverage LLM reasoning via OpenAI and Anthropic APIs.

You will implement layout-aware parsing, multi-column extraction, and table handling, delivering clean JSON or relational outputs.

Formación

• Hands-on experience building OCR or document extraction pipelines in production.
• Strong Python skills with clean, maintainable code and error handling.
• Practical experience with FastAPI (routing, DI, async, middleware).
• Prompt engineering experience with OpenAI or Anthropic APIs, designing extraction chains.

Responsabilidades

• Design and build end-to-end PDF/document extraction pipelines yielding structured output.
• Choose and implement OCR strategies per document type (native text, layout-aware, image-based).
• Parse complex layouts including multi-column text, tables, and forms.
• Output clean JSON or relational data from raw documents.
• Develop and maintain FastAPI services for processing and batching.

Conocimientos

Python

FastAPI

OCR pipelines

Prompt engineering

Herramientas

pdfplumber

PyMuPDF

pdfminer

Tesseract

EasyOCR

AWS Textract

Google Document AI

Azure Form Recognizer

LangChain

LlamaIndex

Descripción del empleo

Offshorly is seeking an AI Developer to build OCR-powered document processing pipelines that extract structured data from PDFs and image inputs. You will own ingestion, OCR, parsing, and API delivery with FastAPI, designing robust extraction chains that leverage LLM reasoning via OpenAI and Anthropic APIs.

You will implement layout-aware parsing, multi-column extraction, and table handling, delivering clean JSON or relational outputs.