Infrastructure Optimization & AI Technical Lead

KeyBank•Brooklyn, OH

8d•$96,000 - $181,000

About The Position

The Infrastructure Optimization & AI (IOAI) organization is transforming Enterprise Technology Services from reactive operations and manual analysis toward proactive, predictive intelligence and governed generative AI adoption. Its Operational Digital Intelligence Platform (ODP) supports conversational intelligence products including AskIT, Ask for Network, and AskIT for NOC. The Technical Lead bridges enterprise architecture and hands-on software development. The role defines and implements scalable Google Cloud Platform (GCP) architectures, writes and reviews production code, and serves as a technical authority for platform development. The lead guides cross-functional engineering squads delivering data pipelines, deterministic policy engines, indexing and retrieval systems, and Google Enterprise AI agent workflows. The role includes designing modular cloud architectures, establishing canonical schemas and reusable patterns, translating architecture and product plans into engineering milestones, mentoring engineers, and developing Python microservices on Cloud Run. It also covers deterministic validation engines, operational telemetry contracts, resilient pipelines, agentic AI orchestration, hybrid retrieval and GraphRAG, evidence provenance, AI evaluation and safety gates, CI/CD, security and access governance, observability, and GCP cost efficiency. The position is based at 4910 Tiedeman Road, Brooklyn, Ohio. The company prioritizes in-office presence while offering flexible options when work can be performed effectively in a mobile environment. The posting is marked #LI-Remote. The annual base salary range is $96,000 to $181,000, with eligibility for incentive compensation. Benefits are referenced through a link, but no specific benefits are listed in the posting.

Requirements

At least 5 years of professional software development experience.
Deep proficiency in Python, including modern asynchronous patterns, typing, and Pydantic v2.
At least 4 years of hands-on GCP architecture and engineering experience.
Experience with GCP serverless compute and orchestration, including Cloud Run, Cloud Functions, Pub/Sub, and Cloud Scheduler.
Experience with BigQuery, including partitioning, clustering, and policy tags; Cloud Storage; and Cloud Spanner.
Experience with GCP security and IAM services, including Secret Manager, IAM roles and service accounts, and workload identity federation.
Direct experience architecting applications using Google Vertex AI, including Gemini models, Agent Builder, Search, grounding pipelines, and vector embeddings, or frameworks such as LangChain or LangGraph.
Demonstrated experience designing distributed systems, RESTful APIs, event-driven pipelines, and canonical data models.
Proven ability to lead squads, mentor developers, define architectural standards, and build consensus across engineering, product, and platform leadership.
Bachelor’s degree in Computer Science, Software Engineering, or Mathematics, or equivalent practical enterprise experience.

Nice To Haves

Google Cloud Certified Professional Cloud Architect or Professional Data Engineer certification.
Familiarity with enterprise network concepts such as firewalls and CIDR routing, ServiceNow ITSM or knowledge management, or IT Operations Analytics.
Experience building automated LLM evaluation harnesses, prompt regression suites, or synthetic evaluation datasets.
Experience with enterprise governance, model risk management, and regulatory compliance in a financial institution.
Experience with Spanner Graph is a strong plus.

Responsibilities

Design and implement modular, loosely coupled GCP cloud architectures that scale independently and meet enterprise reliability and zero-trust standards.
Establish canonical schemas, interface contracts, and reusable patterns for ingestion, normalization, storage, and AI reasoning.
Translate product roadmaps and Architecture Review Board specifications into engineering milestones, technical spikes, and interface contracts.
Mentor squad developers and partner engineering teams, provide code reviews, and give architectural guidance across sprints.
Develop high-performance, containerized Python microservices and jobs using FastAPI and Pydantic v2 on GCP Cloud Run.
Build deterministic validation and evaluation engines that isolate critical business logic from probabilistic LLM outputs.
Implement strict schema validation and data contracts for structured and unstructured operational telemetry.
Design idempotent pipelines, quarantine routing for malformed payloads, and self-healing deployment models.
Architect conversational agent workflows using Vertex AI Agent Builder, Gemini, and agentic frameworks, with custom tool routing, memory, and complexity routing.
Implement hybrid retrieval using Spanner Graph, vector embeddings, and Vertex AI Search for grounded, multi-hop operational reasoning.
Ensure AI-generated responses include verifiable evidence citations linked to authoritative rule IDs, commit hashes, or documentation.
Partner with QA and AI Assurance teams to build evaluation pipelines that assess grounding and latency and enforce safety gates.
Implement automated testing, container scanning, and zero-touch deployment pipelines for repeatable, immutable cloud releases.
Enforce ABAC, pre-retrieval entitlement filtering, Secret Manager integration, and Model Armor safety filters.
Integrate end-to-end tracing through Cloud Logging and Cloud Monitoring, balancing performance with GCP cost efficiency.