AI Engineer/Developer

Il y a 5 heures

Brussels, Brussels, Belgique MEERAB GROUP Temps plein
pulliDesign and implement AI-powered applications using LLMs, RAG, and agentic workflow patterns. /liliBuild backend services, APIs, and integrations needed for AI-centric applications. /liliDesign and improve retrieval pipelines, including chunking, embeddings, vector search, hybrid search, metadata filtering, and reranking. /liliWork with LLM APIs and AI orchestration frameworks to build reliable workflows. /liliImplement evaluation, testing, monitoring, and observability for AI applications. /liliHelp define safe and practical patterns for tool use, human-in-the-loop approval, and agentic behavior. /liliCollaborate with other engineers and product/business stakeholders. /liliSupport model and framework selection based on quality, cost, latency, maintainability, and security. /liliHelp troubleshoot issues such as hallucinations, poor retrieval quality, high latency, cost inefficiency, and unreliable outputs.br/ /li /ulpRole: AI Engineer/Developer /ppLocation: brussels BE /ppLangugae: French/Dutch /pulliDesign and implement AI-powered applications using LLMs, RAG, and agentic workflow patterns. /liliBuild backend services, APIs, and integrations needed for AI-centric applications. /liliDesign and improve retrieval pipelines, including chunking, embeddings, vector search, hybrid search, metadata filtering, and reranking. /liliWork with LLM APIs and AI orchestration frameworks to build reliable workflows. /liliImplement evaluation, testing, monitoring, and observability for AI applications. /liliHelp define safe and practical patterns for tool use, human-in-the-loop approval, and agentic behavior. /liliCollaborate with other engineers and product/business stakeholders. /liliSupport model and framework selection based on quality, cost, latency, maintainability, and security. /liliHelp troubleshoot issues such as hallucinations, poor retrieval quality, high latency, cost inefficiency, and unreliable outputs.br/ /li /ulh3Requirements /h3ulliYou are able to speak, read and write fluently English and French or Dutch. /liliMinimum 5 years of experience as Software Engineer, preferably in backend or full-stack development. /liliMinimum 1 to 2 years of experience integrating LLMs or other generative AI services into software applications. /liliPractical knowledge of RAG, embeddings, vector search, and retrieval quality improvement. /liliStrong programming skills in Python. /liliExperience with MCP, A2A, tool calling, or multi-agent workflows. /liliExperience designing maintainable services with proper testing, logging, CI/CD, and deployment practices. /liliUnderstanding of AI application evaluation, including test datasets, quality metrics, regression testing, and user feedback. /liliGood understanding of cloud-native application development. /liliSecurity-conscious mindset when working with internal or sensitive data. /liliAbility to communicate technical trade-offs clearly and pragmatically. /liliAbility to distinguish when AI, RAG, or agents are useful and when a simpler deterministic solution is better. /liliNice to Have /liliExperience with AI frameworks such as LangGraph, LangChain, Semantic Kernel. /liliExperience with observability or evaluation tools. /liliFamiliarity with Azure cloud environment. /liliExperience with Microsoft AI ecosystem tools such as Semantic Kernel, Microsoft /liliAgent Framework, Microsoft Foundry, or Microsoft 365 Agents SDK. /liliExperience with .
NET / C#. /liliExperience with vector databases or enterprise search platforms. /liliExperience working in regulated, security-sensitive, or enterprise environments. /liliThe right consultant should be comfortable working across software engineering, AI integration, evaluation, security, and production readiness. They should be able to move fast when building MVPs, while still applying enough engineering discipline to make successful solutions scalable and maintainable. /li /ul /p