AI Specialist III

We are one of the largest electric power holding companies in the United States, providing electricity to 7.7 million retail customers in six states. We have approximately 51,000 megawatts of electric generating capacity in the Carolinas, the Midwest and Florida – and natural gas distribution services serving more than 1.6 million customers in Ohio, Kentucky, Tennessee and the Carolinas.

We are transforming our customers’ experience, modernizing our energy grid, generating cleaner energy and expanding our natural gas infrastructure to create a smarter energy future for our customers.

As part of the Digital Transformation organization, you will be part of a fast-paced, entrepreneurial, energetic, and high performing team helping to design cloud native applications and services to enhance the Customer experience. This group is nimble and creative and will shape our technology roadmap. You will be focused on developing well-designed, efficient code that adheres to the latest industry standards and assist in coding, testing, and debugging new software or making enhancements to existing software for customers.

You will be working with designers, developers, and product owners in a small team setting to integrate software components into fully functional software systems spanning multiple business areas.

The Digital Transformation team is located at the
*** Innovation Center at Optimist Hall with product teams delivering a wide-range of leading edge solutions. Read more about the Innovation Center at Optimist Hall.

• Design and implement agentic AI solutions using AWS Bedrock, including Agents, Knowledge Bases, and Guardrails
• Build serverless architectures using AWS Lambda, Step Functions, and Event Bridge to orchestrate AI agent workflows
• Design and engineer sophisticated prompts for agentic AI systems that can plan, reason, and execute multi-step workflows
• Develop prompt architectures that enable AI agents to use tools, APIs, and external resources effectively
• Create and refine chain-of-thought, tree-of-thought, and ReAct (Reasoning + Acting) prompting strategies
• Build evaluation frameworks to measure agent performance, reliability, and safety
• Collaborate with ML and software engineers to integrate prompt strategies with agent frameworks (Lang Chain, Bedrock Agent Core) and leveraging Bedrock agents/services/foundational models
• Optimize prompts for accuracy, efficiency, cost, and latency in production environments
• Implement guardrails and safety measures to ensure responsible agent behavior
• Document prompt patterns, best practices, and design principles for the broader team
• Stay current with the latest research in agentic AI, prompt engineering, and LLM capabilities
• Design and build highly complex software solutions (specifically 2+ years of experience with AWS serverless development and 4+ years of experience on a programming language such as Java or Python)
• Program well-designed, efficient code independently
• Collaborate with designers, developers, and product owners to build solutions
• Clearly communicate team wide progress and identify blocking processes/patterns that hold a team back
• Participate in paired programming and code reviews with juniors to expand the team’s collective knowledge and understanding of software engineering
• Complete independent research to refine best practices and present work to peers
• Turn complex ideas into manageable pieces of work to complete as part of a product team
• Coach and mentor prompt engineers
• Provide technical guidance to leadership so that solutions are built with innovation and sustainability in mind
• Learn the business drivers of a product and help drive the technical direction of a product to fulfill the desired vision

• 3+ years of experience in prompt engineering, AI/ML, or related fields
• Deep understanding of large language models (e.g. Claude) and their capabilities
• Proven experience designing prompts for autonomous agents and multi-step reasoning tasks
• Strong knowledge of agentic frameworks and patterns (ReAct, Plan-and-Execute, Reflection, etc.)
• Strong experience with AWS Bedrock (Agents, Knowledge Bases, Guardrails, and…