Senior Agentic AI Engineer

Verizon•Basking Ridge, NJ

4d•Hybrid

About The Position

As a Senior Agentic AI Engineer within Verizon's team, you will be at the forefront of the company's transformation toward centralized intelligence and enterprise-grade AI agents that drive action. This role manages a broad portfolio of business processes underpinned by analytics and will lead the evolution from reactive reporting to autonomous, intelligent execution. You will architect and deploy multi-agent systems capable of reasoning, planning, and collaborating across complex business domains, aiming to close the loop between AI investment and measurable business outcomes.

Requirements

Bachelor's degree or four or more years of work experience.
Four or more years of relevant experience required, demonstrated through one or a combination of work and/or military experience, or specialized training.
Four or more years of relevant work experience, with two or more years focused specifically on Generative AI, LLMs, and autonomous agent systems.
Four or more years of experience developing and implementing analytical or AI solutions to complex business problems.
Hands-on proficiency with multi-agent orchestration frameworks: LangChain, LangGraph, Google Agent Development Kit (ADK), LlamaIndex, or AutoGen.
Experience designing agent-to-agent (A2A) communication protocols, task delegation hierarchies, and the Model Context Protocol (MCP) for enterprise tool integration.
Experience with agent memory and state management: short-term context, long-term storage, context window optimization, and stateful workflow execution.
Demonstrated knowledge of retrieval-augmented generation (RAG), ONNX-based embeddings, vector databases (Pinecone, Weaviate, pgvector), and semantic search.
Demonstrated knowledge of cloud-scale data engineering: BigQuery pipelines and GCP (Cloud Run, Vertex AI, Amazon Bedrock, or Azure equivalent).
Experience with Python and SQL for statistical modeling, prompt engineering, token management, and large-scale data extraction.

Nice To Haves

A Master's degree in a quantitative discipline such as Data Science, Computer Science, Statistics, Mathematics, Engineering, or Operations Research.
Familiarity with ML/LLM monitoring and observability tooling (e.g., OpenTelemetry, Arize Phoenix, Galileo) for tracking model drift, accuracy degradation, and agent output quality over time.
Experience with generative AI-augmented analytics: AI testing frameworks, model validation pipelines, and LLMOps tooling including observability and token management.
Experience with AI governance, security, compliance, and financial ROI modeling for AI agent deployments.
Domain expertise in customer, churn prediction, customer lifetime value (CLV) modeling, or related commercial analytics in Consumer, Telecommunications, Financial Services, or Technology industries.
A passion for educating and communicating AI findings with integrity to all levels—from reviewing raw outputs with colleagues to presenting AI strategy and business impact to executive stakeholders.
High curiosity, an investigative mindset, and the flexibility to adapt to new challenges while staying focused on the team's deliverables.

Responsibilities

Architecting and deploying scalable multi-agent AI systems—designing agent workflows, communication protocols (A2A), and task delegation hierarchies for fleets of cooperative AI agents.
Integrating LLMs with enterprise systems via APIs, function calling, and the Model Context Protocol (MCP) to enable agents that can reason, plan, use tools, and act autonomously.
Building and operating enterprise Agent Factory pipelines—moving from reactive, episodic analytics to autonomous, continuously executing intelligent workflows.
Designing and implementing retrieval-augmented generation (RAG) systems with ONNX-based embeddings and vector databases to power context-aware, hyper-personalized outputs at scale.
Managing agent memory and state architecture—implementing short-term, medium-term, and long-term memory using tools like LangGraph for reliable multi-step task execution.
Owning the full agent and model lifecycle: feature engineering, training, production deployment (Cloud Run, Vertex AI), and continuous monitoring with observability tooling.
Establishing security guardrails, permission boundaries, short-lived agent identity tokens, and safety constraints for production agentic systems.
Developing and maintaining propensity models, classification systems, and forecasting solutions that feed downstream agent decision-making.
Translating complex agentic AI outputs into clear operational strategies and presenting findings to senior leadership and cross-functional business partners.
Performing advanced AI-augmented analytics and generative AI model validation to ensure accuracy, reliability, and measurable business performance.