Software Engineer II – Agentic AI Engineer

Citizens•Iselin, NJ

15h•Hybrid

About The Position

The Software Engineer II – Agentic AI Engineer is a mid‑level engineer responsible for building, integrating, and operating agent‑based AI capabilities within production software systems. The role focuses on implementing autonomous or semi‑autonomous AI agents that interact with tools, services, and workflows while adhering to established architectural patterns, security controls, and software engineering best practices. This role operates with moderate autonomy, delivers independently on scoped components, and collaborates closely with senior engineers, product managers, and platform teams.

Requirements

Strong software engineering fundamentals (data structures, APIs, distributed systems basics)
Proficiency in one or more backend languages (e.g., Python, Java)
Hands‑on experience with LLMs, prompt engineering, or agent frameworks
Familiarity with REST APIs, message queues, and cloud‑native patterns
Ability to independently deliver well‑defined features and components
Strong debugging and problem‑solving skills
Clear communication of technical concepts to cross‑functional partners
Willingness to learn rapidly evolving AI and agentic patterns
Typically 3–6 years of professional software engineering experience
Prior exposure to AI/ML systems, LLMs, or workflow automation is expected

Nice To Haves

Experience operating production systems (on‑call, incident support) is a plus

Responsibilities

Design and implement AI agents that can reason, plan, and act across defined tasks and workflows
Develop prompt structures, memory handling, tool invocation logic, and agent orchestration patterns
Implement agent guardrails including input validation, response constraints, and fail‑safe behaviors
Apply responsible AI practices (traceability, explainability, and bias awareness) in agent design
Build production‑ready services that embed AI agents into applications and workflows
Write clean, modular, testable code following team and enterprise coding standards
Implement APIs, background jobs, or event‑driven components that leverage AI agents
Participate actively in code reviews and design discussions
Integrate agentic AI with internal systems such as APIs, data services, workflow engines, and UI layers
Implement secure access patterns for agent tool usage (authentication, authorization, secrets handling)
Ensure agents operate within defined system boundaries and business rules
Support deployment of agentic AI components in cloud or containerized environments
Implement logging, monitoring, and basic telemetry for agent behavior and outcomes
Assist in identifying agent failure modes, hallucination risks, and recovery strategies
Contribute to runbooks and operational documentation
Develop unit and integration tests for agent logic, prompts, and decision paths
Validate agent behavior across normal, edge, and failure scenarios
Participate in performance tuning to reduce latency and cost of agent execution
Work closely with product and business partners to translate requirements into agent behaviors
Collaborate with data, platform, and security teams to align on standards and constraints
Provide technical input into sprint planning and estimation
Share learnings and patterns with peers through documentation and demos