AI Engineer – LLM / Agentic Systems (f/m/x)
⚲ Warszawa, Wrocław, Kraków, Katowice, Gdańsk, Poznań, Białystok, Bydgoszcz, Łódź, Lublin
Do uzgodnienia
Wymagania
- Python
- LLM
- OpenAI
- LangChain / LangGraph / LlamaIndex
- SQL
- vector databases
Opis stanowiska
We are building a new AI team within a project in the digital healthcare / data-driven marketing domain. We are looking for both a Tech Lead and Senior Engineers who will work together as one cohesive team.
The role focuses on designing and delivering production-grade AI systems powered by LLMs in a real-world, regulated environment.
Your tasks
• Design and develop production-ready AI systems based on LLMs and agent-based workflows
• Build and maintain data pipelines for retrieval, orchestration, and evaluation
• Implement solutions using frameworks such as LangChain, LangGraph, LlamaIndex, or custom approaches
• Develop and optimize RAG pipelines (chunking, embedding, retrieval, reranking)
• Integrate LLM models (OpenAI and open-source) and support model selection decisions
• Collaborate with Data Science teams to productionize ML models
• Build evaluation frameworks (metrics, test sets, regression tests) and monitor system performance
• Create and maintain scalable AI infrastructure in a cloud environment
Requirements
• Minimum 5 years of commercial experience in software development or AI/ML engineering
• Strong Python skills with solid software engineering practices (modular design, testing, code reviews)
• Hands-on experience with LLM frameworks such as LangChain, LangGraph, or LlamaIndex
• Experience with vector databases (e.g., Pinecone, Weaviate, pgvector, OpenSearch)
• Practical experience building RAG pipelines and working with unstructured data
• Experience working with SQL and cloud-based data systems
• Availability to start within 2 weeks due to a fast project kickoff
• Fluency in English
• Residing in Poland required
The role focuses on designing and delivering production-grade AI systems powered by LLMs in a real-world, regulated environment.
Your tasks
• Design and develop production-ready AI systems based on LLMs and agent-based workflows
• Build and maintain data pipelines for retrieval, orchestration, and evaluation
• Implement solutions using frameworks such as LangChain, LangGraph, LlamaIndex, or custom approaches
• Develop and optimize RAG pipelines (chunking, embedding, retrieval, reranking)
• Integrate LLM models (OpenAI and open-source) and support model selection decisions
• Collaborate with Data Science teams to productionize ML models
• Build evaluation frameworks (metrics, test sets, regression tests) and monitor system performance
• Create and maintain scalable AI infrastructure in a cloud environment
Requirements
• Minimum 5 years of commercial experience in software development or AI/ML engineering
• Strong Python skills with solid software engineering practices (modular design, testing, code reviews)
• Hands-on experience with LLM frameworks such as LangChain, LangGraph, or LlamaIndex
• Experience with vector databases (e.g., Pinecone, Weaviate, pgvector, OpenSearch)
• Practical experience building RAG pipelines and working with unstructured data
• Experience working with SQL and cloud-based data systems
• Availability to start within 2 weeks due to a fast project kickoff
• Fluency in English
• Residing in Poland required
🔍 Dekoder Ogłoszenia
🔴
We are building a new AI team within a project in the digital healthcare / data-driven marketing domain.
Zespół jest nowy i dopiero powstaje w ramach istniejącego projektu, co może oznaczać niepewność co do jego długoterminowej stabilności i zakresu działania.
🔴
We are looking for both a Tech Lead and Senior Engineers who will work together as one cohesive team.
Oznacza to, że zespół będzie miał silną hierarchię i prawdopodobnie Tech Lead będzie miał decydujący głos w kwestiach technicznych, co może ograniczać autonomię Senior Engineerów.
🔴
The role focuses on designing and delivering production-grade AI systems powered by LLMs in a real-world, regulated environment.
Praca w środowisku regulowanym (np. medycznym) może oznaczać dodatkowe, czasochłonne procesy certyfikacji i zgodności, które nie są typowe dla standardowych projektów IT.
🔴
production-grade AI systems
Wymaga to nie tylko stworzenia działającego prototypu, ale także zapewnienia wysokiej dostępności, skalowalności, bezpieczeństwa i monitorowania, co jest znacznie bardziej wymagające niż typowe zadania badawcze.
🔴
Collaborate with Data Science teams to productionize ML models
Może oznaczać, że inżynierowie AI będą musieli zajmować się nie tylko rozwojem modeli, ale także ich wdrażaniem i utrzymaniem w środowisku produkcyjnym, co jest zadaniem często pomijanym w opisach pracy.