AI Data Engineer: Real-Time RAG & Vector Infra

Kyndryl Inc.; Comunidad de Madrid; Hybrid; EUR 70.000 - 100.000; Vollzeit

Zusammenfassung

Kyndryl is seeking an AI Data Engineer to architect high-performance real-time data pipelines that empower autonomous systems. You will design RAG pipelines, scale vector databases, and transform unstructured IT logs into actionable embeddings, while ensuring data quality and safety.

You will build semantic layers, automate guardrails for noise, bias and PII, and develop CI/CD pipelines to keep context windows fresh as production evolves.

Qualifikationen

• Experience in data mining, storage and ETL processes.
• Experience building data pipelines with Glue, Databricks, Synapse or Dataproc.
• Experience with relational and NoSQL databases like PostgreSQL, DB2, MongoDB.
• Strong problem-solving, analytical and critical thinking.
• Ability to manage multiple projects with attention to detail.
• Ability to translate technical requirements from business needs.

Aufgaben

• Architect RAG pipelines for real-time data processing.
• Scale vector databases to ensure sub-second retrieval.
• Build semantic layers and knowledge graphs for context.
• Automate data guardrails to detect noise, bias, or PII.
• Develop CI/CD pipelines for data infrastructure to keep context windows fresh.

Kenntnisse

Data mining

ETL processes

Data pipelines

Relational databases

NoSQL databases

Analytical skills

Multi-project management

Communication

Ausbildung

Degree in Computer Science / Software Engineering / IT

Open Certified Technical Specialist - Data Engineering

AWS Certified Data Analytics – Specialty

Google Cloud Professional Data Engineer

Microsoft Certified: Azure Data Engineer Associate

Elastic Certified Engineer

Tools

AWS Glue

Databricks

Azure Synapse

Google Dataproc

GitHub

Visual Studio

Jobbeschreibung

Kyndryl is seeking an AI Data Engineer to architect high-performance real-time data pipelines that empower autonomous systems. You will design RAG pipelines, scale vector databases, and transform unstructured IT logs into actionable embeddings, while ensuring data quality and safety.

You will build semantic layers, automate guardrails for noise, bias and PII, and develop CI/CD pipelines to keep context windows fresh as production evolves.