Ability to collaborate closely with machine learning engineers and data scientists to refine AI models, develop robust model embeddings, and optimize prompts for accuracy, performance, and cost-efficiency
Develop and optimize AI agent systems, including multi-agent orchestration and tool integration
Implement and iterate on RAG pipelines, evaluation frameworks, and LLM-powered data processing solutions
Maintain and enhance applications and solutions on Python
Requirements
A solid knowledge of Python, with experience in writing clean, efficient, and maintainable code
Ability to design and develop autonomous agents using frameworks such as LangGraph/CrewAI/OpenAI Agents SDK, enabling intelligent decision-making and task automation
Experience in rapid prototyping using large language models (LLMs), vector databases, and knowledge graphs to explore and implement new AI-driven capabilities
Expertise in retrieval-augmented generation (RAG) workflows and document retrieval techniques to enhance content relevance
Hands-on experience working with large-scale language models, such as GPT/Claude/Gemini/Qwen, along with a solid understanding of prompt engineering, embedding techniques
Familiarity with the Model Context Protocol (MCP) for connecting LLM agents to external tools and data sources
Experience in rapid prototyping of AI-driven capabilities using LLMs, embeddings, and structured/unstructured data sources
Experience with code generation and AI-assisted development tools (Claude Code, GitHub Copilot, Cursor, Codex)
Strong written and spoken English communication skills (B2)
Nice to have
Experience with Python frameworks such as Django, Flask, or FastAPI
Proficiency in SQL for working with structured databases, querying data, and optimizing database performance
Familiarity with platforms Dataiku, Snowflake, Tableau, Streamlit, Artifactory, dbt
Hands-on experience with NLP deep learning frameworks such as PyTorch, TensorFlow, and Transformers
Familiarity with graph RAG approaches (Microsoft GraphRAG, LightRAG, or Neo4j-based solutions)
Understanding AI cost optimization: model routing, caching strategies (semantic caching), prompt compression, and token management
