数据工程师/分析师

We are looking for a Senior Data / Ingestion Engineer for our client, a consultancy that helps banks and insurers transform their operations through AI, automation, and advanced analytics, with a strong focus on Anti-Financial Crime (fraud prevention, AML, KYC) and enterprise AI platforms. You will join their delivery for a Dutch insurance client.
We're looking for an engineer with deep experience building robust ingestion pipelines for unstructured documents and integrating OCR and document extraction technologies. You will turn high volumes of insurance documents (PDFs, scans, emails, Office files) into structured, high-quality, AI-ready data that feeds downstream RAG and AI models.
This is a long-term remote-first contract position for candidates based in Europe.
Responsibilities

• Design and implement scalable pipelines for ingesting high-volume unstructured insurance documents (PDFs, scans, emails, Word, Excel, PowerPoint).

• Build connectors to document sources such as SharePoint and email.

• Integrate, configure, and optimise OCR and document parsing technologies to extract high-accuracy text and layouts.

• Build automated workflows for text cleaning, normalisation, semantic chunking, and metadata tagging.

• Design vector storage schemas and robust retrieval mechanisms (RAG) to feed downstream AI models.

• Ensure document processing pipelines meet enterprise security and low-latency SLA requirements.

• Build automated error monitoring and extraction validation loops that flag low-confidence OCR outputs.

• Apply engineering best practices across the pipeline lifecycle: version control, CI/CD, and testing.

Work Conditions

• Start Date: ASAP

• Location: Remote within Europe (CEE preferred)

• Long-term contract-based role: until July 2027 with possible extension

• Contract with EU LCC

Requirements

• 5–10 years' experience in data engineering.

• Proven experience building data processing and document ingestion pipelines on public cloud platforms.

• Hands-on experience processing unstructured documents (PDF, Word, Excel, PowerPoint, scans, emails).

• Experience building connectors to enterprise sources such as SharePoint and email.

• Practical experience with document extraction / OCR tools, e.g. AWS Textract or equivalent.

• Strong engineering practices: Git, CI/CD, automated testing.

• Experience in banking or insurance is an advantage.

Technical Requirements
Mandatory

• Python, SQL

• AWS: S3, Step Functions, CloudWatch

• OCR / document extraction: AWS Textract or equivalent

• Unstructured document processing and ingestion pipelines

• Git, CI/CD, testing

Nice to have

• Vector databases and RAG architectures

• Azure, Databricks

• Financial services / insurance domain experience

Highlights
If you are passionate about turning unstructured documents into AI-ready data, and want to help leading financial institutions fight financial crime with AI, this role is for you!
Originally posted on Himalayas