AI Architect

We are looking for an AI Architect who genuinely gets excited about building things with AI — someone who shipped something real with LLMs, not just played with the API. You will work on international projects for clients in banking, insurance, and telco (US, Netherlands, UK), translating messy business problems into working AI systems.   About Devapo   At Devapo, we focus on continuous self-development and acquiring new knowledge. If you are a fast learner, want to participate in international projects, are a team player, and can work independently - join us! We provide our clients with more than just code - we want to equip them with tools that allow their businesses to flourish. Our clients' success is our success, which is why we ensure that everyone who creates Devapo has a long-term goal in mind. At Devapo, you'll have the opportunity to discuss your challenges and solutions with our team of experts, including experienced architects who are always ready to share their knowledge and guide you through complex technical decisions.   Key Responsibilities   ● Designing GenAI and LLM-based architectures for enterprise use cases — translating business requirements into concrete technical designs  ● Building and delivering POCs and MVPs quickly, then evolving them into production-grade systems  ● Designing RAG pipelines: chunking strategies, embedding, vector indexing, retrieval quality  ● Implementing and reviewing agentic workflows (LangChain, LangGraph, CrewAI or similar)  ● Integrating LLMs (OpenAI, Anthropic, Gemini, open-source) into existing enterprise stacks — ERP, CRM, BPM, data platforms  ● Creating reusable architectural patterns and engineering standards that scale across projects  ● Working directly with clients: understanding requirements, presenting solutions, supporting presales and RFP responses  ● Writing Architecture Decision Records and technical documentation for both technical and business audiences  ● Mentoring engineering teams   Requirements   ● Proven experience building LLM-based systems that run in production — POC to production, not just experimentation ● Solid Python skills and hands-on experience with the AI stack: LangChain, LangGraph, vector databases (Qdrant, Weaviate, Pinecone, pgvector or similar) ● Working knowledge of RAG — you understand why naive chunking fails and what to do about it ● Experience designing AI system architecture: integration patterns, scalability, security, cost trade-offs ● Ability to translate business requirements into technical designs and communicate them clearly to both engineers and stakeholders ● Familiarity with cloud AI services: Azure AI / AWS Bedrock / GCP Vertex ● You know how to ship: REST APIs, Docker, cloud basics ● Awareness of AI governance, responsible AI practices, and regulatory context (EU AI Act, GDPR, DORA) ● English B2+ — client-facing role, calls and written communication included   Nice to Have   ● Experience evaluating LLM outputs (RAGAS, LangSmith, Arize or similar)  ● MLflow or another experiment tracking tool  ● Databricks, Azure AI Foundry or AWS Bedrock  ● Fine-tuning experience (LoRA, PEFT, anything hands-on)  ● Kafka or streaming pipelines for real-time AI use cases   What We Offer   ● Certifications and training funded  ● Private medical care (Medicover)  ● Multisport card  ● English language classes  ● Flexible working hours  ● Team meetups and integration events  ● Referral bonus