Agent Engineer

Dice is the leading career destination for tech experts at every stage of their careers. Our client, RightTalents, is seeking the following. Apply via Dice today!Job Title: Agent EngineerLocation: New York, NY ( Hybrid)Duration: 12 MonthsWork Hours: 40 hrs/weekJob Summary: The Agent Engineer designs, develops, and iterates on AI agents that can reason, plan, and execute multi-step tasks autonomously. This is the core builder role of the agentic technology stack combining software engineering, LLM expertise, and systems thinking to create reliable agents that integrate seamlessly with enterprise tools, APIs, and data sources.Responsibilities:Design and implement AI agents with well-defined goals, tools, memory systems, and reasoning loopsBuild agent tooling integrations — connecting agents to APIs, databases, internal services, and external platforms via function calling or MCPDevelop and maintain agent memory architectures (short-term, long-term, episodic) for context persistence across sessionsImplement robust error handling, retry logic, and fallback behaviors to ensure agent reliability in edge casesWrite comprehensive unit and integration tests for agent behaviors, including adversarial test casesCollaborate with Product Managers to translate business requirements into agent capabilities and constraintsOptimize agent performance — reducing token usage, latency, and cost without sacrificing output qualityParticipate in agent evaluation cycles, reviewing outputs and incorporating feedback into agent designDocument agent architectures, tool schemas, and design decisions for cross-team visibilityStay current with the rapidly evolving agentic AI ecosystem (frameworks, models, protocols, best practices)Required Qualifications:3+ years of software engineering experience; 1+ years working with LLMs or AI agentsStrong proficiency in Python; familiarity with TypeScript/JavaScript a plusExperience with at least one agentic framework: LangChain, Lang Graph, Crew AI, AutoGen, or custom implementationsSolid understanding of LLM capabilities and limitations (context windows, tool use, structured outputs, hallucination patterns)Experience with REST APIs, vector databases (Pinecone, Weaviate, Chroma), and data retrieval systemsFamiliarity with prompt engineering principles and context management strategiesUnderstanding of async programming, event-driven architectures, and distributed systemsStrong debugging skills and comfort with ambiguous, non-deterministic systemsPreferred Skills:Experience fine-tuning LLMs or working with model APIs (Anthropic, OpenAI, Mistral, local models)Background in RAG (Retrieval-Augmented Generation) architecture designContributions to open-source AI/agent frameworks