AI Software Development Engineer

Job Details: Job Description: Build and scale production-grade data infrastructure for agentic AI systems that execute complex, multi-step work with autonomy, state, memory, and tool use. You will engineer resilient platforms for long-running agent workflows, multi-agent coordination, and adaptive execution in enterprise environments - while delivering the data pipelines and integrations that connect agents to enterprise data, legacy systems, and simulation tools. This role emphasizes reliability, control, observability, data quality, and governance for agentic AI systems over conversational chatbot patterns. Key Responsibilities Agentic Orchestration • Productionize graph-based orchestration for planner-executor-validator, orchestrator-worker, and similar patterns. • Implement explicit state and control flows: branching, loops, routing, interruption points, and human approval checkpoints. • Enable robust agent-tool integration across APIs, services, data systems, and enterprise platforms. • Support multi-agent collaboration patterns with guardrails for coordination, delegation, and convergence. Data Engineering and Integration • Design and maintain data pipelines connecting distributed enterprise data to a centralized semantic/knowledge layer that ensures clean, unified inputs for agent consumption. • Build and operate event streaming, API management, and systems integration infrastructure to enable trustable, consistent data for agentic workflows. • Build and maintain a data catalog and onboarding guides for teams adopting the agentic platform. DevOps and Reliability for AI Agent Systems • Define and track SLIs/SLOs for task completion reliability, reasoning quality, tool-call success, latency, and cost across agent pipelines. • Implement CI/CD practices tailored for agent deployments - versioning agent configurations, prompts, tools, and orchestration logic as code. • Build incident response and reliability practices for autonomous workflows, including safe rollback, pause/resume, and controlled retries. • Optimize compute, storage, and inference paths for sustained agent throughput and cost efficiency. Observability, Evaluation, and Control • Implement full-stack observability for agent runs - traces, state transitions, tool telemetry, data quality signals, outcomes, and replay ability. • Build continuous evaluation pipelines for agent behavior, including correctness, safety, drift, and regression detection. • Provide actionable operational dashboards for quality, reliability, data health, and cost in production agent systems. Qualifications: Minimum qualifications are required to be initially considered for this position. Preferred qualifications are in addition to the minimum requirements and are considered a plus factor in identifying top candidates. Minimum Qualifications Master’s degree in software engineering, Computer Engineering, Information Technology, or related field with 5 years of experience. OR PhD in Software Engineering, Computer Engineering, Information Technology, or related field with 3 years of experience. Experience listed above should be in at least one of the following: DevOps, SRE, data engineering, or infrastructure engineering for production AI or distributed systems. LLM serving, retrieval infrastructure, and runtime control for non-deterministic systems. Graph-based or agent orchestration frameworks. Experience building RAG and knowledge-graph-backed systems for LLM applications in production. Python skills for orchestration, data pipeline development, and platform automation Preferred Qualifications • Experience designing multi-agent systems with clear autonomy boundaries and human-in-the-loop controls. • Track record in production evaluation frameworks for agent quality and safety. • Experience with observability and reliability for data and agent pipelines (metrics, logging, tracing, data quality monitoring). • Strong experience with Kubernetes, infrastructure as code, CI/ ... (truncated, view full listing at source)