About this role

- PLACE OF WORK

- 4024 Debrecen, Barna utca 23 (HU-HNL-002)

- AREA OF EMPLOYMENT

- IT

- START OF WORK

- as soon as possible

- EMPLOYMENT TYPE

- Full-time

megállapodás szerint

My responsibilities:

- Design, build, and operate conversational analytics assistants that let business users ask questions of company data in natural language and receive accurate, traceable answers

- Develop LLM-based applications and agents in Python, including text-to-SQL, tool calling, and retrieval over documents and business definitions

- Build and maintain evaluation pipelines, including test sets, automated regression checks, and answer quality monitoring, so every prompt, model, or data change is measured before release

- Ensure solutions are secure and compliant, with permissions enforced in the backend, protection against prompt injection, and responsible handling of personal data under GDPR

- Monitor and optimize production systems for accuracy, latency, and cost, including model selection, caching, and support for multiple languages

- Work closely with data engineers, analysts, and business stakeholders to turn requirements into reliable solutions, and document all technical processes to ensure operational continuity

The knowledge I own:

- Strong programming knowledge in Python, including building and testing production services and APIs

- Hands-on experience shipping LLM-based applications to production using managed model APIs

- Strong SQL knowledge and experience with text-to-SQL or other natural language interfaces over structured data

- Strong understanding of retrieval-augmented generation (document parsing, chunking, hybrid search, reranking)

- Experience evaluating LLM systems (building test sets, regression testing, using LLM-as-a-judge and knowing its limits)

- Good understanding of agent design, tool calling, structured outputs, and context management

- Basic understanding of LLM security (prompt injection, access control) and data protection requirements (GDPR)

- Basic understanding of data modelling principles and semantic layers (metric definitions, business glossaries)

- Pragmatic, data-centric mindset: chooses the simplest solution that works and measures before optimizing

- Structured, analytical thinking

- Strong communication skills, including explaining what AI can and cannot do to non-technical stakeholders

- Well-organized

- Practical experience with Databricks Genie, Microsoft Fabric AI Agents or SAP Joule is appreciated

The offer that would convince me:

- A constantly growing organization and increasing opportunities

- Secure, long-term job opportunity

- Varied and engaging job responsibilities

- The chance to build AI products used by colleagues across the international organization

- Outstanding salary

- Flexible work arrangements

- Home office possibility

Online application:

Please use our online application and attach your resume.

AIIS Adatkezelési tájékoztató

Privacy notice