Lead LLM Platform Engineer with Bedrock, Vertex AI, and Agentic Workflows

Bank

Columbus OH

Work onsite 4 days per week in Columbus

Needed ASAP

Contract to hire

Must work on W2

Job Title

Lead Solution & Deployment Engineer LLM Platform (Bedrock, Vertex AI, Agentic Workflows)

Job Description

We re hiring a senior-level engineer to lead the deployment and enablement of large language models (LLMs) across cloud platforms like AWS Bedrock and Google Vertex AI. This is a hands-on engineering role focused on designing secure, auditable, and reusable LLM delivery patterns with emphasis on agentic workflows, prompt orchestration, observability, and guardrail enforcement.

You ll work across teams from cyber and risk to architecture and product to build platform-level capabilities that accelerate the safe, compliant adoption of LLMs for high-impact use cases. You ll be responsible for converting proof-of-concepts into production-ready deployments, instrumented for telemetry, policy enforcement, and post-deployment validation.

This role is for someone who has a strong grasp of how LLMs behave in real-world applications, not just in notebooks and knows how to design systems that are both flexible and trusted.

Responsibilities

• Design and implement reusable deployment blueprints for LLMs across Bedrock and Vertex AI, with support for both batch and real-time inference.
• Build infrastructure to manage prompts, chains, retrieval strategies (RAG), and memory in a modular, auditable way.
• Enable secure use of foundation models, including usage quotas, red-teaming sandboxes, API gateways, and runtime policy enforcement.
• Partner with architecture and cyber teams to embed controls aligned with internal AI governance and regulatory expectations.
• Implement observability layers that track model behavior, prompt performance, drift, cost, and data leakage risks.
• Guide application teams in integrating LLMs into downstream products via APIs, workflows, and agents.

Required Qualifications

• 6+ years in engineering roles, including experience with LLM application delivery in a cloud environment.
• Hands-on experience deploying models on AWS Bedrock, Vertex AI, or similar managed foundation model platforms.
• Proficiency with Python and frameworks such as LangChain, LlamaIndex, or orchestration libraries like Ray, Argo, or Prefect.
• Deep understanding of prompt tuning, embedding models, retrieval workflows, and agentic decision trees.
• Experience implementing security, telemetry, or access control for ML or GenAI applications in enterprise environments.

Preferred Qualifications

• Experience implementing runtime validators or using tools like Guardrails AI, PromptLayer, or LlamaGuard.
• Familiarity with SR 11-7, ISO 42001, or internal AI governance frameworks.
• Experience instrumenting LLM apps with usage-based cost controls, token metering, and model behavior audits.
• Knowledge of secure software supply chain practices for GenAI use cases (e.g., prompt versioning, output sandboxing, inference isolation).