AI Engineer

What You Will Do

Your responsibilities will include

• Designing and building production-ready RAG pipelines — document ingestion, chunking strategies, embedding generation, vector storage, retrieval logic, and response generation using LLMs
• Integrating LLM APIs into existing product backends — OpenAI, Anthropic, Mistral, or open-source models — with proper error handling, retry logic, streaming support, and cost awareness
• Working with vector databases including Pinecone, Weaviate, Qdrant, or pgvector — managing indexes, optimizing retrieval, and ensuring search quality meets product requirements
• Implementing evaluation frameworks to measure model output quality — building evals, tracking regressions, and creating feedback loops that improve system performance over time
• Designing and maintaining AI-related APIs that integrate cleanly with the rest of the product stack — authentication, rate limiting, input validation, and output formatting all included
• Building and maintaining LLM orchestration workflows using frameworks such as LangChain or LlamaIndex — chaining prompts, managing context windows, and handling multi-step reasoning tasks
• Collaborating with backend and frontend engineers to integrate AI features into existing product surfaces — you understand the full stack well enough to work effectively across it

Requirements The following are non-negotiable for this role:

Technical

• 2 to 5 years of overall engineering experience with at least 1 to 2 years working specifically with LLMs or AI systems in a production environment — not just experimentation or coursework
• Strong Python skills — you write clean, maintainable, production-quality Python; you understand dependency management, async patterns, and testing
• Hands-on experience with at least one major LLM provider — OpenAI, Anthropic, Mistral — and familiarity with open-source alternatives such as Llama or Falcon
• Practical experience building RAG systems — you have designed the full pipeline, not just the retrieval step
• Working knowledge of vector databases and embedding models — you have used them in production and understand their tradeoffs
• Ability to evaluate model outputs systematically — you do not rely on vibes; you build evals, track metrics, and iterate based on data
• Solid API design skills — AI features exist within larger systems and you understand how to design the interfaces that connect them cleanly
• Experience with LangChain, LlamaIndex, or a comparable orchestration framework in a real project context

Availability & Setup

• Full-time availability: 40 hours per week with no parallel full-time commitments
• Minimum 4 hours of consistent daily overlap with Central European Time
• Stable internet connection and a reliable, dedicated home office setup

Communication

• English proficiency at B2 level or above — you communicate directly with a European client engineering team in meetings, written updates, and technical documentation
• Able to explain AI system behavior, limitations, and tradeoffs clearly to both technical and non-technical stakeholders