Software Developer (AI)

CDQ Poland is a company focused on innovative Data Quality Services, leveraging Data Sharing as a method to improve data quality. They provide a cloud-based software platform offering Data Quality as a Service, continuously monitoring and updating customer master data, integrated with ERP or CRM systems.

• English language proficiency minimum B2 + Polish or C1
• Minimum 3 years of experience
• Java: Spring Boot, Docker
• AI-related: Spring AI
• Experience integrating LLMs into applications (OpenAI API, Anthropic, local inference, etc.)
• Understanding of vector databases (Milvus, Pinecone, Qdrant, Elasticsearch vector search, or similar)
• Knowledge of AWS Bedrock
• Knowledge of LangChain4j
• Knowledge of embeddings, prompt engineering basics, and retrieval-augmented generation (RAG)
• Understanding of Model Context Protocol

Nice to have:

• Redis
• MongoDB
• Experience with AI/ML frameworks or orchestration libraries (LangChain4j, Spring AI, Embabel, Haystack, etc.)
• Familiarity with LLM model lifecycle: prompt design, evaluation, latency considerations, cost/performance trade-offs
• Experience with Ollama / vLLM
• Experience with streaming architectures for AI pipelines (i.e. Kafka Streams)
• Experience with document processing, OCR, or semantic search

• Designing and implementing AI agents with reasoning pipelines (e.g. multi-step workflows, tool usage, RAG-based decision making)
• Integrating AI capabilities such as LLM-powered services, semantic search, and intelligent automation
• Contributing to scalable architectures for data- and event-driven systems
• Improving, refactoring, and maintaining existing codebases
• Designing tasks in collaboration with the Team Lead and Product Owner
• Participating in code reviews, architecture discussions, and knowledge sharing
• Contribute to the design of scalable architectures for data-heavy and AI-driven services
• Collaborate closely with the Team Lead and Product Owner to design tasks and shape technical solutions
• Participate in code reviews, architecture discussions, and knowledge-sharing sessions