Project & Program Manager

Project & Program Manager
Location: Preferred location NJ or Texas. Open to remote

JD:

The Role in One Sentence
You sit at the intersection of marketing, product, and engineering - translating business goals for campaign generation into clear technical deliverables, managing cross-functional execution across AI/ML and data teams, and ensuring the marketing operations is optimized to expedite campaign planning and execution efforts and significantly improve the Go-To-Market speed.

The Marketing Operations AI Agent stack includes:

• LangChain / LangGraph for orchestrating multi-step agent workflows
• Elastic as the vector database powering semantic retrieval and campaign content search
• Data ingestion pipelines that feed brand assets, campaign history, product data, and audience segments into the agent
• Prompt engineering layers that translate business campaign briefs into structured LLM instructions
• LLM APIs (OpenAI / Gemini) for voice handling, campaign management
• Future integration with other marketing operation tools in the ecosystem such as Adobe, Workato, and more.

But components don't upgrade or maintain the AI Agent. Someone needs to orchestrate and manage the process from requirements to final AI Agent releases, and drive every workstream - from data ingestion to UAT - to a production-ready outcome. That person is the Technical Program Manager.

What You'll Do
Primary Responsibilities

Business Engagement: Partner with marketing operation leaders and managers to translate operation needs into clear requirements the engineering team, for both new releases and ongoing AI Agent support, can execute against.

Roadmap & Delivery Management: Own the program roadmap for the AI Agent. Break epics into sprints, manage dependencies across product design, AI engineering, data engineering, and QA, and keep delivery on track.

Technical Coordination: Drive execution across LangGraph workflow development, Elastic vector DB configuration, ingestion pipeline builds, and prompt engineering iterations. You don't architect - you coordinate, unblock, and track.

Stakeholder Communication: Translate technical progress and trade-offs into language the marketing stakeholders understand. Translate business feedback into actionable tickets for engineering. Run demos, reviews, and status reports.

Data Pipeline Oversight: Coordinate ingestion pipeline delivery with data engineering. Ensure brand assets, product catalogs, campaign history, and audience data are flowing correctly into Elastic indexes on schedule.

Risk & Issue Management: Proactively surface blockers - model hallucination spikes, data freshness gaps, latency issues, stakeholder misalignment - and drive resolution before they become launch blockers.

Adoption & Enablement: Help marketing teams get value from the agent. Run onboarding sessions, document workflows, gather structured feedback, and feed it back into the roadmap.

Ongoing AI Agent Support: Ensure AI Agent is well kept to support day to day operations: monitor and manage necessary updates and upgrades ranging from infrastructure changes, issue fixes, enhancements, and quality refinement.

Day-in-the-Life Examples

Working with Stakeholders: on a daily basis, work with the Marketing Operations team to align on AI Agent issues impacting operations, prioritize and determine urgent and lower priority fixes. On a monthly basis,

1. review AI Agent product roadmap to ensure full alignment of marketing operations priorities. Refine release plans accordingly;
2. review field feedback to determine enhancements and refinements needed for the AI agent. Refine release plan accordingly.

Working with Engineering teams: monitor daily standups and focused work sessions to ensure all work streams are on track to secure agreed upon release plans and commitments to issue troubleshooting. Hands-on to triage and expedite issue resolution, roadblock removal, and quality reassurance to ensure zero disruption to the day to day marketing operations.

Working with Transformation Leadership team: ensure all adoption and organization transformation needs are properly supported by the engineering team on time and with superb quality, to ensure zero-disruption to the marketing transformation initiatives.

Ongoing: participate in strategic planning efforts along with the client partner to support mid- and long-term marketing operations goals and priorities.

What You Won't Do

• Design the LangGraph agent architecture - that's the AI engineering team
• Configure Elastic indexes or write ingestion pipeline code - that's data engineering
• Make model selection decisions - that's the AI platform team
• Manage a team of engineers - you're a delivery and coordination force-multiplier

Required Qualifications
Technical Fluency (Must-Have)
You don't need to build these systems. You need to understand them well enough to ask the right questions, recognize when something is wrong, and communicate trade-offs to non-technical stakeholders.

LangGraph / LangChain: Understand agent workflows, nodes, edges, tool use, memory, and state. Can read a flow diagram and ask informed questions. Why it matters: You'll coordinate LangGraph development sprints and triage issues in agent behavior.

LLM Basics: How LLMs generate text, what prompts do, what hallucination is, how temperature and context windows affect outputs. Why it matters: You'll translate stakeholder quality feedback into technical root causes - prompt vs. data vs. model.

Prompt Engineering: Understand prompt structure, system vs. user messages, few-shot examples, output formatting. Can review a prompt and give meaningful feedback. Why it matters: You'll manage the prompt library and govern prompt versioning and evaluation.

Elastic / Vector DB: Understand what a vector database does, what embeddings are, how semantic search differs from keyword search. Not required to write queries. Why it matters: You'll track ingestion pipeline delivery and understand retrieval quality issues.

Data Ingestion Pipelines: Understand pipeline stages - source extraction, transformation, loading - and what can go wrong. Not a data engineer, but can speak the language. Why it matters: You'll coordinate data engineering dependencies and unblock data freshness issues.

Evaluation & Quality Metrics: Understand how to define quality for AI outputs - what makes a campaign 'good', how to measure it, what regression detection means. Why it matters: You'll own eval governance and translate marketing feedback into measurable quality signals.

Domain skills needs to be familiar with digital and user facing product management, in addition to AI fluency. You will own delivery of an agentic driven digital platform

Program Management Skills (Must-Have)

Delivery ownership: You've owned complex, multi-team programs from kickoff to production. You know how to manage scope, schedule, dependencies, and risk simultaneously.

Agile/Scrum fluency: You can run sprint planning, backlog grooming, retros, and standups. You know when to push back on scope and when to escalate.

Requirements translation: You turn ambiguous business asks into structured engineering requirements. You can write a ticket that engineers can act on without a 45-minute meeting.

Stakeholder management: You manage up to marketing leadership, across to engineering leads, and down into day-to-day execution without losing any thread.

Issue triage mindset: When something breaks or slips, you diagnose fast, communicate clearly, and drive resolution - not just escalate and wait.

Business Acumen (Equally Critical)

Marketing domain literacy: You understand how campaigns are built - briefs, channel strategy, copy iteration, brand guidelines, approval workflows. You don't need to be a marketer, but you need to speak the language.

Consultative mindset: The first request from marketing ('make the agent write better ads') is rarely the real problem. You ask the next question.

Communication clarity: Superb communication skills for a marketing organization whose first language is not IT or technology. The right culture fit.

Bias to shipping: Pilots and demos are easy. You're measured on agents in production, not slides in a deck.

Experience

• 5 8 years in technical program management, product management, or engineering delivery roles
• At least one full life cycle implementation of programs that include AI/ML, LLM-based, or data-intensive products in production
• Demonstrated experience working with marketing, growth, or content teams as primary stakeholders
• Track record of shipping complex, multi-dependency programs on schedule in an agile environment
• Experience in marketing technology, adtech, e-commerce, or a consumer-facing product organization strongly preferred