Principal Engineer - Software Development

Verizon•Basking Ridge, NJ

3d•Hybrid

About The Position

At Verizon Corporate Security, we are driving innovation and completely transforming how we build software to protect the world's most reliable network. We’ve proven that by pairing precise business requirements with AI-assisted development, we can compress delivery timelines by an order of magnitude—delivering the promise of the digital world faster than ever. To scale this capability across our organization, we are bringing on a Principal Automation Engineer: AI & Agentic Workflows. This is not a role for someone who wants to manage from the sidelines. This is an opportunity for a highly autonomous, hands-on builder to join the V Team. If you have a "founding engineer" mindset, you will thrive here—turning written concepts into working, high-impact production software that unlocks immediate business value and helps us move the world forward.

Requirements

Bachelor’s degree or four or more years of work experience.
Six or more years of relevant experience required, demonstrated through one or a combination of work and/or military experience, or specialized training.
5+ years of experience building and shipping production software end-to-end, demonstrating clear personal ownership, identifying specific users impacted, and delivering measurable business outcomes.
Experience shipping Large Language Model (LLM) applications to end users, applying hands-on expertise in prompt and context engineering, retrieval mechanisms, agent design, and system evaluation.
Demonstrated experience integrating data across complex, imperfect systems, leveraging strong SQL skills, practical data agility, and deep comfort navigating cloud environments (GCP) and business intelligence tools (Looker or equivalent).

Nice To Haves

10+ years of relevant engineering experience building, shipping, and scaling enterprise software.
A Bachelor's degree in Information Technology, Information Systems, Computer Science, or a related technical field.
Direct experience working alongside business operators in a highly embedded capacity, such as a forward-deployed engineer, solutions engineer, or founding engineer.
Exposure to regulated-data domains, including telecom CPNI, eDiscovery, financial compliance, or healthcare.
Proven experience operationalizing and scaling AI systems in production environments, with a strong emphasis on system feedback loops and continuous improvement.
Experience working with physical security, Operational Technology (OT), or critical-infrastructure technology.
A verifiable track record operating in high-consequence environments, securely handling highly sensitive information (such as law enforcement data or executive location information).
Exceptional "spec literacy", including the proven experience reading business processes, asking targeted probing questions to surface hidden edge cases, and drafting precise technical specifications.
A proven track record of operating with high autonomy, consistently clearing your own blockers without a program-management layer and escalating issues by proposing solutions rather than just reporting status updates.

Responsibilities

Build applications from specifications: Take completed business requirements specifications and deliver working software against them as a hands-on individual contributor (write code, configure platforms, wire data, and demo results).
Build and operate the AI toolchain: Own internal tooling making the spec-to-build loop repeatable, including structured intake requirements-interview agents, reusable prompt/context/evaluation scaffolding, and build telemetry instrumentation.
Scale Automation Through Agentic Workflows: Design, build, and deploy agents on the Kairos platform, leveraging the enterprise ecosystem to ensure consistency, reusability, rapid iteration, and maximum speed-to-value.
Build necessary data plumbing: Integrate against imperfect upstream systems (SQL Server, BigQuery/EDW, GCP services, Looker, vendor APIs), managing schema drift, partial coverage, and inconsistent definitions while ensuring proper data lineage and governance.
Prove it works before it is trusted: Build evaluation harnesses alongside features (shadow-mode testing, precision/recall measurement, override-feedback loops, human-in-the-loop approval paths).
Demo, ship, and iterate weekly: Ship high-value features to staging weekly and demo to stakeholders on a standing cadence while navigating operational and policy complexity.