Role Overview
Oracle AI Data Platform (AIDP) seeks a Platform Software Engineer to build enterprise AI agents and production ML services. You will work across Deep Insights (natural-language data exploration, SQL, visualizations, agent-driven research) and MLOps (experiment tracking, model versioning, deployment). The role combines hands-on AI engineering with reliable cloud service development, including Python services, Agent Runtimes, MLflow integration, asynchronous execution, streaming responses, tenant isolation, and observability.
Responsibilities
- Design, build, and operate backend services for Deep Insights agents and MLOps within AIDP.
- Develop conversational analytics connecting natural-language questions to enterprise data, SQL execution, visualizations, and research findings.
- Integrate Agent Runtimes and tools supporting streaming responses, conversation history, and protocols such as MCP and A2A.
- Improve agent accuracy, relevance, and transparency through metadata enrichment, evaluation, and user feedback.
- Build MLflow-compatible capabilities for experiment tracking, model registration, versioning, and artifact management.
- Develop model deployment workflows covering packaging, activation, updates, rollback, and operational visibility.
- Design secure APIs enforcing user identity, authorization, and customer/workspace data isolation.
- Improve service performance, scalability, and resilience through testing, metrics, distributed tracing, and production troubleshooting.
- Collaborate with engineering, product, and data science teams to deliver production-ready capabilities.
Requirements
- Strong proficiency in Python and API Development.
- Experience with MLflow or similar experiment tracking and model management platforms.
- Solid understanding of SQL, data access patterns, and tenant isolation.
- Familiarity with Agent Runtimes, asynchronous execution, and streaming responses.
- Knowledge of secure API design, authorization, and observability (metrics, distributed tracing).
- Ability to collaborate across engineering, product, and data science teams.