Lead Software QA Engineer, AI Automation

Zeta Global

About The Position

Zeta Global is seeking a Lead Software QA Engineer, AI Automation to join their AI-first engineering organization. This role is crucial for enabling developers to ship fast, confidently, and continuously by owning the system of how quality, data, and delivery come together. This is a high ownership role where you will shape how an AI-First team builds, validates, and ships production software, moving beyond traditional QA and project management. The position involves making quality a built-in system, defining what is buildable early in the process, owning the data reality, bringing domain depth in MarTech/AdTech and healthcare, and driving delivery like an owner. You will redefine QA for AI-First Development by building testing strategies for AI-generated and non-deterministic systems, validating outputs, and creating evaluation frameworks for AI reliability, accuracy, and behavior. The ideal candidate will think in systems, be comfortable pushing back on Product and Sales, and know how to make engineers better without becoming a bottleneck.

Requirements

Likely outgrown traditional QA or delivery roles
Thinks in systems, not tickets
Can go from data model → product requirement → test strategy → delivery plan without handoffs
Comfortable pushing back on Product and Sales when things do not make sense
Knows how to make engineers better without becoming a bottleneck
Cares deeply about what happens in production, not just what gets shipped
Deep experience in quality engineering, SDET, or technical delivery leadership
Strong hands-on experience with test automation and CI/CD systems
Ability to query and reason across complex data systems (SQL required)
Experience with AI-assisted development or ML/LLM-based systems
Strong familiarity with MarTech / AdTech ecosystems
Healthcare experience is a major plus

Responsibilities

Make Quality a Built-In System, Not a Gate
Turn quality into developer behavior, not a downstream function
Teach engineers to anticipate failure modes, edge cases, and data issues before writing code
Build automated test systems that act as guardrails, not overhead
Define what “production-ready” actually means and enforce it through systems, not process
Eliminate “hope-driven releases”
Define What’s Buildable Before It Gets Built
Partner with Sales and Product to pressure test ideas early
Translate ambiguity into clear execution plans engineers can run with
Kill bad ideas early. Shape good ones into something buildable
Own the Data Reality
Map and understand how data actually flows, not how people think it flows
Identify what data exists, what is missing, and what is unreliable
Navigate internal systems and third-party integrations with confidence
Ensure every feature is grounded in real, usable data, not assumptions
Bring Domain Depth (MarTech / AdTech + Healthcare)
Apply real-world understanding of identity, activation, measurement, and data constraints
Operate within healthcare realities, including compliance and sensitivity of data
Ensure what we build is not just technically correct, but market-relevant and viable
Drive Delivery Like an Owner
Run delivery for a team of AI-first engineers
Break down work into clear, executable steps with no ambiguity
Drive predictable delivery without slowing the team down
Surface risks early and adjust before they become problems
Align Product, Engineering, and GTM without friction
Redefine QA for AI-First Development
Build testing strategies for AI-generated and non-deterministic systems
Validate outputs, not just code
Create evaluation frameworks for AI reliability, accuracy, and behavior
Ensure AI accelerates development without degrading quality