Senior Software Engineer, Agentic Technologies

- Design, develop, and manage agentic systems that effectively perform real-world tasks beyond answering questions - Architect systems that support planning, memory, tool usage, and multi-step execution, choosing suitable patterns based on specific constraints and risks - Maintain a balance between autonomy and control by creating predictable, debuggable, secure agents aligned with business objectives - Build agents using contemporary frameworks such as LangGraph, LangChain, or AutoGen, incorporating structured outputs and state management - Design and implement MCP and/or RAG-based integrations for agents to access tools, data, and context - Ensure security, consent, and access control for all interactions between agents and tools, collaborating with platform teams to establish integration standards - Integrate agentic systems into current services and platforms via APIs, ensuring production readiness from start to finish - Define metrics for success in agent performance (accuracy, success rate, latency) to promote continuous improvement - Create automated evaluation pipelines and instrument agents for debugging and iterative development in live environments - Design fallback and recovery mechanisms for failure scenarios, proactively mitigating potential risks - Establish architectural standards and best practices for agent development, enhancing technical quality through reviews, documentation, and mentorship