Grain Slovakia s.r.o
We are looking for a
Senior AI Engineer , expert in Python focused on AI/ML, NLP, or LLM-based applications. For our client, we are enriching our existing development team responsible for further developing one of our most successful products. The product is an integrated delivery platform that leverages real-world engagement experience, enabling cross-team collaboration, real-time transparency, and insightful decision-making. Responsibilities: Architect end-to-end LLM-based pipelines using RAG, embeddings, and orchestration layers. Design multi-agent systems using LangChain, LangGraph, CrewAI, or AutoGen, emphasizing modularity and scalability. Evaluate model options (e.g., OpenAI, Azure OpenAI) based on performance, cost, and capabilities. Define embedding strategies, context windows, and prompt structures for complex reasoning tasks. Build chunking and ingestion pipelines for PDFs and unstructured documents using Python SDKs (LangChain, pyPDF, RecursiveCharacterTextSplitter). Integrate and deploy vector stores (e.g., Azure AI Search, FAISS, Postgres, ChromaDB) with semantic search and reranking techniques. Implement CI/CD pipelines, test coverage, and API integrations (e.g., Postman, GitHub Actions). Build and maintain reusable Python classes for LLM tools, pipelines, agents, and evaluation modules. Debug and optimize complex distributed Python systems involving multiple services and third-party APIs. Develop and maintain Jupyter Notebooks for prototyping, analysis, and storytelling. Create and deploy Docker containers for reproducible environments. Work with CI/CD pipelines to ensure high-quality, testable code. Use coding tools like Cursor and GitHub Copilot to accelerate development while ensuring security and reusability. Develop and orchestrate autonomous agents with well-defined roles, tools, and memory-sharing strategies. Implement observer and fallback agents to enhance system resilience and reduce hallucinations. Use telemetry and observability tools (e.g., DataDog, LangFuse) for monitoring and debugging. Define and track evaluation metrics for GenAI and RAG systems (faithfulness, precision, recall, F1, semantic similarity). Use sklearn, NumPy, and RAGAS evaluation loops to validate model performance. Implement output schema validation, prompt constraints, and quality assurance for enterprise readiness. Build interactive prototypes using Streamlit, LangFlow, or Jupyter Notebooks. Translate complex models into clear, explainable insights for stakeholders. Lead design sessions and mentor developers on AI best practices. Prerequisites and skills: 8+ years in software engineering, with 3+ years focused on AI/ML, NLP, or LLM-based applications. Expert in Python, with experience building modular, production-grade systems. Proven experience with GenAI solutions, prompt engineering, embedding retrieval, and multi-agent orchestration. Hands-on experience with LangChain, LangGraph, CrewAI, AutoGen, Hugging Face. Experience processing large unstructured data (PDFs, HTML, JSON) with LangChain and embedding models. Proficient with vector databases like Azure AI Search, FAISS, Postgres with pgvector. Knowledge of Jupyter, Docker, GitHub, REST APIs. Familiar with DevOps and CI/CD pipelines. Skilled in evaluating LLM and RAG pipelines using relevant metrics and tools like sklearn, NumPy, RAGAS. Experience with observability tools such as LangFuse and DataDog. Comfortable with AI-powered developer tools like GitHub Copilot and Cursor. Preferred qualifications: Experience with multimodal LLMs, vision-language models, or tool-augmented inference. Understanding of reasoning models, ReAct prompting, and planner-executor agents. Experience with agile methodologies and cross-functional teams. Ability to evaluate and compare RAG pipelines using structured and human-in-the-loop methods. Strong communication skills for technical and executive audiences. Further information: Senior level, fully remote, with occasional in-person sessions in Prague. US hours overlap (2-6 pm CET), preferably starting at 2 pm. Language: EN. Start: ASAP. About the company:
Technology and consulting software company. Apply for the position:
Submit your first name, last name, email, phone, optional introduction, and resume (max 20 MB in DOC, DOCX, ODT, PDF, RTF, or TXT formats).
#J-18808-Ljbffr
Senior AI Engineer , expert in Python focused on AI/ML, NLP, or LLM-based applications. For our client, we are enriching our existing development team responsible for further developing one of our most successful products. The product is an integrated delivery platform that leverages real-world engagement experience, enabling cross-team collaboration, real-time transparency, and insightful decision-making. Responsibilities: Architect end-to-end LLM-based pipelines using RAG, embeddings, and orchestration layers. Design multi-agent systems using LangChain, LangGraph, CrewAI, or AutoGen, emphasizing modularity and scalability. Evaluate model options (e.g., OpenAI, Azure OpenAI) based on performance, cost, and capabilities. Define embedding strategies, context windows, and prompt structures for complex reasoning tasks. Build chunking and ingestion pipelines for PDFs and unstructured documents using Python SDKs (LangChain, pyPDF, RecursiveCharacterTextSplitter). Integrate and deploy vector stores (e.g., Azure AI Search, FAISS, Postgres, ChromaDB) with semantic search and reranking techniques. Implement CI/CD pipelines, test coverage, and API integrations (e.g., Postman, GitHub Actions). Build and maintain reusable Python classes for LLM tools, pipelines, agents, and evaluation modules. Debug and optimize complex distributed Python systems involving multiple services and third-party APIs. Develop and maintain Jupyter Notebooks for prototyping, analysis, and storytelling. Create and deploy Docker containers for reproducible environments. Work with CI/CD pipelines to ensure high-quality, testable code. Use coding tools like Cursor and GitHub Copilot to accelerate development while ensuring security and reusability. Develop and orchestrate autonomous agents with well-defined roles, tools, and memory-sharing strategies. Implement observer and fallback agents to enhance system resilience and reduce hallucinations. Use telemetry and observability tools (e.g., DataDog, LangFuse) for monitoring and debugging. Define and track evaluation metrics for GenAI and RAG systems (faithfulness, precision, recall, F1, semantic similarity). Use sklearn, NumPy, and RAGAS evaluation loops to validate model performance. Implement output schema validation, prompt constraints, and quality assurance for enterprise readiness. Build interactive prototypes using Streamlit, LangFlow, or Jupyter Notebooks. Translate complex models into clear, explainable insights for stakeholders. Lead design sessions and mentor developers on AI best practices. Prerequisites and skills: 8+ years in software engineering, with 3+ years focused on AI/ML, NLP, or LLM-based applications. Expert in Python, with experience building modular, production-grade systems. Proven experience with GenAI solutions, prompt engineering, embedding retrieval, and multi-agent orchestration. Hands-on experience with LangChain, LangGraph, CrewAI, AutoGen, Hugging Face. Experience processing large unstructured data (PDFs, HTML, JSON) with LangChain and embedding models. Proficient with vector databases like Azure AI Search, FAISS, Postgres with pgvector. Knowledge of Jupyter, Docker, GitHub, REST APIs. Familiar with DevOps and CI/CD pipelines. Skilled in evaluating LLM and RAG pipelines using relevant metrics and tools like sklearn, NumPy, RAGAS. Experience with observability tools such as LangFuse and DataDog. Comfortable with AI-powered developer tools like GitHub Copilot and Cursor. Preferred qualifications: Experience with multimodal LLMs, vision-language models, or tool-augmented inference. Understanding of reasoning models, ReAct prompting, and planner-executor agents. Experience with agile methodologies and cross-functional teams. Ability to evaluate and compare RAG pipelines using structured and human-in-the-loop methods. Strong communication skills for technical and executive audiences. Further information: Senior level, fully remote, with occasional in-person sessions in Prague. US hours overlap (2-6 pm CET), preferably starting at 2 pm. Language: EN. Start: ASAP. About the company:
Technology and consulting software company. Apply for the position:
Submit your first name, last name, email, phone, optional introduction, and resume (max 20 MB in DOC, DOCX, ODT, PDF, RTF, or TXT formats).
#J-18808-Ljbffr