GenAI Python Developer
DCV Technologies
⚲ Wrocław
Wymagania
- Git Workflows
- Asynchronous and streaming data pipelines
- APIs and AI services
- Python Developer
- Relational databases and SQL
- AI / GenAI systems
- SQL databases
Opis stanowiska
We are looking for a GenAI Python Developer to join a team building AI-powered solutions within enterprise environments. The role focuses on developing intelligent AI systems, integrating APIs, and implementing modern GenAI workflows using Python. You will work on designing and implementing agentic AI architectures, integrating large language models, and building scalable backend systems that power AI-driven applications. Send CV to (marcillina.tietjen@dcvtechnologies.co.uk) if you are interested. Location Wroclaw, Poland – Hybrid (minimum 2 days onsite) Employment Type Contract / B2B Key Responsibilities • Develop AI-driven applications using Python • Design and implement GenAI workflows and AI agents • Integrate external APIs and AI services • Work with asynchronous and streaming data pipelines • Collaborate with engineering teams in Agile/Scrum environments • Implement AI orchestration and workflow management • Maintain high-quality code using Git workflows • Support data integration and backend services using SQL databases Required Skills • 5+ years Python development experience • Hands-on experience building AI / GenAI systems • Experience with agentic AI architectures (autonomous agents, planning, tool usage, memory) • Experience calling APIs and handling async/streaming data • Good understanding of Git workflows (branches, PRs, merge conflict resolution) • Basic understanding of relational databases and SQL • Strong collaboration and communication skills • Ability to work hybrid in Wroclaw (2 days onsite) Nice to Have • Experience with LangChain, LangGraph or similar AI frameworks • Experience with vector databases and embeddings • Knowledge of prompt engineering • Familiarity with Azure DevOps pipelines (CI/CD) • Experience working with multi-agent systems • Experience with memory management in LLM architectures