AI Software Engineer Semantic Data & LLM - 20384
Veritaz AB · Stockholm · 6 månader – upp till 12 månader · Deadline 2026-11-25
Veritaz AB seeks an **AI Software Engineer** specialising in semantic data and large language models for an engineering assignment on a data platform team. The position involves developing and maintaining a **governed semantic layer** that bridges structured data warehouse systems and unstructured document sources. Core duties include engineering LLM-driven agents, building pipelines for document parsing and entity extraction, and implementing embedding-based Retrieval-Augmented Generation (RAG) workflows. The work also centres on establishing reliable, production-grade retrieval architectures that maintain strict source traceability and respect data access permissions. Candidates need a degree in Computer Science, Data Science, Mathematics, or equivalent practical experience, alongside at least five years of experience building and operating tested production software. **Strong skills in Python or TypeScript** and robust SQL abilities with relational data models are essential, as is practical experience with LLM agents, RAG, and unstructured data processing techniques. Familiarity with semantic layers like Malloy or dbt, knowledge graphs, Google Cloud Platform, Snowflake, or data governance frameworks is considered advantageous.
Veritaz is a leading IT staffing solutions provider in Sweden, committed to advancing individual careers and aiding employers in ensuring the perfect talent fit. With a proven track record of successful partnerships with top companies, we have rapidly grown our presence in the USA, Europe, and Sweden as a dependable and trusted resource within the IT industry.
Job Overview
Role Summary
An experienced AI Software Engineer is needed to join a data platform team focused on semantic modelling, large language models (LLMs), and unstructured data. In this hands-on engineering assignment, you will help extend a governed semantic layer spanning both structured warehouse data and document-based information. The role focuses on building reliable AI-powered data capabilities with traceability, permission-aware retrieval, and production-quality implementation.
Key Responsibilities
- Extend and maintain a governed semantic layer across structured and unstructured data sources.
- Design and build LLM-based agents that combine data warehouse information with document-derived knowledge.
- Develop pipelines for document parsing, information extraction, entity extraction, and entity resolution.
- Create retrieval solutions that preserve source traceability and respect access permissions.
- Implement and evaluate embedding-based retrieval approaches, including Retrieval-Augmented Generation (RAG) solutions.
- Assess and improve the quality, accuracy, and reliability of extraction and retrieval methods.
- Develop, test, and operate robust production systems using well-tested code.
Technical Stack & Requirements
- Strong development skills in Python and/or TypeScript.
- Practical experience with LLM-based solutions, including agents and Retrieval-Augmented Generation (RAG).
- Experience with unstructured-data processing, including document parsing, information extraction, entity resolution, embeddings, and retrieval.
- Strong SQL skills and understanding of relational data models and data warehouses.
- Experience with semantic modelling or semantic layers, such as dbt Semantic Layer or Malloy, is advantageous.
- Knowledge of ontologies or knowledge graphs is advantageous.
- Experience with Google Cloud Platform, Snowflake, or data governance practices is advantageous.
Qualifications & Experience
The ideal candidate has at least 5 years of experience building and operating production systems with a strong focus on code quality and testing. A degree in Computer Science, Data Science, Mathematics, or equivalent practical experience is required. Experience working with semantic data models, knowledge graphs, cloud data platforms, and governed data environments is considered beneficial.