AI Engineer / Application Analyst

4 days in-office / 1 day remote

Designs, builds, and evolves AI solutions that deliver measurable business and operational value across Southern Power.

The AI Engineer translates use-case intent into production-ready agents, models, and workflows, working within the enterprise architectural, governance, and operational standards to ensure solutions are scalable, safe, and fit for a critical-infrastructure environment.

Experience building AI and machine learning applications in enterprise settings.

Practical understanding of agentic AI concepts, including planners, tool-calling, memory, and multi-agent collaboration.

Familiarity with enterprise data platforms and APIs, including structured, unstructured, and document-based data.

Awareness of AI governance, risk, and compliance considerations in regulated environments

Strong problem-solving skills with the ability to decompose complex use cases into modular agent

behaviors.

Product-oriented mindset, focused on delivering tangible outcomes rather than experimentation alone.

Ability to work effectively within defined architectural and governance constraints.

Clear communicator who can explain AI behavior and limitations to non-technical stakeholders.

Proficiency in Python and AI frameworks for LLMs and agents.

Proficiency with Databricks preferred.

with prompt engineering, evaluation techniques, and structured outputs.

Familiarity with vector databases, embeddings, and retrieval-augmented generation (RAG) patterns.

Working knowledge of API integration and tool execution within agent workflows.

Exposure to CI/CD, model deployment, and observability concepts sufficient to operate within pipelines.

Develop AI solutions.

Build and configure agents that reason, plan, and act using models, tools, and enterprise data, aligned to approved patterns.

Implement use-case workflows end-to-end.

Translate business and operational requirements into AI workflows spanning data ingestion, inference, tool execution, and human-in-the-loop review.

Apply responsible AI and guardrails.

Embed safety controls, validation steps, and explainability mechanisms into agent logic and prompts by design.

Collaborate across the AI delivery lifecycle.

Partner with architects on design decisions and with ML Ops to ensure smooth deployment, monitoring, and iteration.

Optimize performance and cost.

Tune prompts, models, and agent interactions to improve accuracy, latency, and cost efficiency.

Support experimentation to production.

Rapidly prototype new AI capabilities while adhering to standards that enable promotion to enterprise-grade deployments.

Continuously improve agent behavior.

Use feedback, evaluations, and operational signals to refine agent performance and reliability over time

Some relocation assistance is available if you do not already live within a reasonable commute distance.