Описание
sa.global Labs develops empower, an industry-specific agentic AI platform that combines a domain knowledge graph, a unified data hub, and multi-agent orchestration to automate enterprise decisions. Its three pillars are Corporate Knowledge, Data Hub, and Multi-Agent Skills.
Задачи
•Design, build, and operate production-grade REST APIs and backend services for empower's Data Hub and Agent Workflows
•Integrate open-source and commercial AI/LLM tools into backend systems, including RAG pipelines, embeddings, vector search, and model routing across providers
•Build and maintain data pipelines and integrations that feed the Business Knowledge Graph
•Own service reliability, performance, security, monitoring, cost, and scaling
•Collaborate with AI Product Engineers to expose platform primitives, including tool endpoints, retrieval APIs, and data contracts, for agent workflows
•Integrate open-source and commercial AI capabilities into the platform foundation reliably
Требования
•At least 5 years of backend software development experience, with Python as the primary language for AI-adjacent work
•Proven experience designing and shipping production APIs and distributed services
•Hands-on experience integrating LLM/AI APIs into real applications
•Working knowledge of retrieval architectures using PostgreSQL with pgvector and pg_search, chunking, and hybrid vector/full-text search
•Familiarity with MCP (Model Context Protocol)
•Experience with cloud-native deployment, containerization, and CI/CD
•Daily hands-on use of Claude Code or a comparable AI coding agent for scaffolding, refactoring, test generation, and codebase navigation
•Demonstrate agency, systemic thinking, structured communication for AI-directed work, ownership and quality discipline, and comfort with ambiguity
Nice to have: Neo4j or other graph databases, MLflow or another experiment-tracking/model-lifecycle tool, Prefect or a comparable orchestrator such as Airflow, LlamaIndex or a comparable RAG framework, knowledge-graph or ontology-backed systems, services-centric domains such as legal, Architecture/Engineering/Construction or ERP, Azure AI, Databricks, or similar enterprise data platforms.
Условия
•Hybrid working model, flexible depending on candidate location
•Role location: Belgrade, Serbia, or Lisboa, Porto, or Braga, Portugal
