Job Summary: Design the knowledge and generative artificial intelligence strategy for agents, including RAG architecture, document preparation, model selection, and cost optimization. Key Highlights: 1. Design the conceptual architecture of the knowledge base for agents. 2. Define the RAG strategy and official content quality standards. 3. Propose criteria for model selection and routing. Join Stefanini! At Stefanini, we are more than 30\.000 brilliant minds, connected from 41 countries, doing what they love and co\-creating a better future. **Responsibilities and Authorities** * Design the knowledge and generative AI strategy for agents, including RAG architecture, document preparation and segmentation, model selection and routing, quality evaluation, performance metrics, cost optimization, and observability mechanisms. * The individual will be responsible for transforming official knowledge sources and use-case requirements into a technically viable architecture for agents and conversational experiences. **Primary Responsibilities** * Design the conceptual architecture of the knowledge base for agents. * Define the RAG strategy, including: * Document ingestion. * Content extraction. * Cleaning and normalization. * Segmentation or chunking. * Enrichment with metadata. * Embedding generation. * Indexing. * Semantic and hybrid retrieval. * Re\-ranking. * Source citation and traceability. * Define quality and structural requirements for official content. * Design versioning, publishing, unpublishing, and document update mechanisms. * Define strategies to differentiate between current, historical, contradictory, or pending-approval content. * Design interactions among the knowledge base, moderator, agents, and language models. * Propose criteria for model selection based on complexity, cost, latency, accuracy, and criticality. * Design routing rules between models and agents. * Define optimization strategies for: * Token consumption. * Latency. * Cost per conversation. * Response reuse. * Cache usage. * Context size and quality. * Define evaluation metrics for responses and retrieval: * Accuracy. * Relevance. * Groundedness. * Faithfulness. * Coverage. * Abstention rate. * Hallucinations. * Latency. * Cost. * Design test cases and scenarios for technical and functional evaluation. * Define guardrails for responses, tools, sensitive data, and agent actions. * Design observability mechanisms for prompts, models, tokens, responses, sources, and errors. * Document risks associated with generative AI usage and model limitations. * Collaborate with the Data Architect to define dependencies and source quality. * Collaborate with the Integration Architect to define capabilities and tools accessible to agents. * Participate in the conceptual design of the moderator and agent orchestration. * Generate inputs for the technical backlog and subsequent implementation phase. **Requirements and Qualifications** **Technical Requirements** * Bachelor’s degree in Systems Engineering, Software Engineering, Computer Science, Artificial Intelligence, or related fields. * Professional experience in data engineering, AI, machine learning, semantic search, or intelligent platform development. * Practical experience in: * LLMs and generative AI applications. * RAG. * Embeddings. * Vector databases. * Semantic and hybrid search. * Prompt engineering. * Model and response evaluation. * Document processing. * Proficiency in Python and AI model APIs. * Experience with vector databases or search engines. * Knowledge of chunking techniques, metadata filtering, reranking, and retrieval. * Ability to design technical and functional quality metrics. * Knowledge of observability, traceability, and cost optimization in AI solutions. * Ability to document architectures, experiments, decisions, and evaluation criteria. **Desirable** * Experience with Azure OpenAI, Azure AI Search, Databricks, Microsoft Fabric, Amazon Bedrock, Vertex AI, OpenSearch, Elasticsearch, Pinecone, Weaviate, or equivalent technologies. * Experience with LangChain, LlamaIndex, Semantic Kernel, or other frameworks. * Knowledge of agents, tool calling, function calling, and multi-agent orchestration. * Experience in OCR and intelligent document processing. * Knowledge of LLM security and OWASP Top 10 for LLM Applications. * Experience in designing virtual assistants or conversational channels. **Additional Information** Are you looking for a place where your ideas shine? With over 38 years of experience and a global presence, at Stefanini we transform tomorrow—together. Here, every action matters, and every idea can make a difference. Join a team that values innovation, respect, and commitment. If you are a disruptive professional, committed to continuous learning, and innovation is in your DNA, then we’re exactly what you’re looking for. Come—let’s build a better future together!
Valentina Rodríguez
Indeed · HR
Valentina Rodríguez
Indeed · HR
