Applied AI Engineer

At ERCOT, our diverse and dynamic work environment provides a platform on which employees can work together to build the future of the Texas power grid and wholesale market utilizing the latest technologies and resources. We encourage you to join our talented, dedicated workforce to develop world-class solutions for today and tomorrow’s energy challenges while learning new skills and growing your career.
ERCOT is committed to fostering inclusion at all levels of our company. It is the cornerstone of our corporate values of accountability, leadership, innovation, trust, and expertise. We know that individuals with a wide variety of talents, ideas, and experiences propel the innovation that drives our success. An inclusive and diverse workforce strengthens us and allows for a collaborative environment to solve the challenges that face our industry today and in the future.
JOB SUMMARY
Applies knowledge of generative AI application development, agent orchestration, retrieval architecture, and evaluation methodology to deliver production AI solutions. Follows established AI governance, security review, and evaluation processes to deploy reliable, auditable systems. Operates in a regulated environment where delivery speed is balanced against required governance and review gates.
JOB DUTIES:
Translates ambiguous business problems into scoped technical roadmaps, and identifies constraints (data access, compliance, latency, cost) before development begins.
Designs and builds production agentic systems, covering planning, tool-calling, multi-step reasoning, memory, and error recovery, using modern orchestration frameworks such as LangGraph or Microsoft Agent Framework.
Implements production RAG pipelines, including chunking, embeddings, hybrid search, reranking, retrieval-quality evaluation, and content freshness.
Builds and extends connectors that give agents secure, standardized access to enterprise tools and data.
Deploys applications onto managed cloud platforms and integrates them with enterprise systems and collaboration tools.
Builds evaluation suites, tracing, and rollback paths so agents are reliable in production rather than demonstrations.
Monitors, debugs, and improves deployed applications against evaluation metrics.
Applies sound system-design principles. Defines application architecture, data flows, and integration boundaries, and designs for scalability, reliability, latency, and cost.
Codifies repeatable patterns, turning successful builds into reusable components and reference architecture the team can leverage.
Works directly with non-technical business owners to understand their workflows, and maintains current knowledge of evolving LLM capabilities, implementation patterns, and AI development stacks.
EXPERIENCE
Requires minimum 5 years job related work experience in AI/ML or software engineering in excess of degree requirements.
Required skills and knowledge:
Proven record of building and deploying production-grade autonomous agents, not prototypes.
Agent orchestration frameworks (LangGraph, Microsoft Agent Framework, or comparable).
Production RAG with vector search and vector databases (pgvector, Azure AI Search, Databricks Vector Search)
Strong Python and hands-on LLM API integration.
System-design fundamentals: scalable, reliable, maintainable services, API and integration-boundary design, and trade-offs across latency, throughput, and cost.
Building or extending tool and data connectors for LLM applications.
Deploying and operating applications on a managed cloud platform.
AI governance, model lifecycle, and evaluation methodology.
Stakeholder and discovery skills. Works directly with non-technical business owners, scopes ambiguity, and operates autonomously.
Preferred:
Solution and system architecture across multiple applications, with security-by-design and reference architecture.
Large-scale data platforms (Databricks) for retrieval, feature, or pipeline work.
Experience in a regulated industry (energy, finance, healthcare) or an audit-driven environment.
Multi-agent orchestration and context engineering.
EDUCATION
Bachelor's Degree: Computer Science, Data Science, Information Systems, Engineering, or related field (Required) or a combination of education and experience that provides equivalent knowledge to a major in such fields is required
TOOLS & TECHNOLOGY
Agent & LLM Frameworks: LangGraph, Microsoft Agent Framework, LangChain, LlamaIndex
LLM Platforms & APIs: Claude API, Azure OpenAI, OpenAI API, model routing and evaluation frameworks
AI Coding Assistants: Claude Code, OpenAI Codex, GitHub Copilot, Microsoft Copilot Studio
Retrieval & Vector Search: Azure AI Search, Databricks Vector Search, pgvector
Data & Analytics: Databricks, Power BI, SQL, Oracle DB, PostgreSQL
Connectors & Integration: MCP (Model Context Protocol), REST APIs, enterprise system connectors, Teams integration
Cloud & Deployment: Azure, OpenShift (Private Cloud / On-Premises), Docker, Kubernetes, Helm
CI/CD & Source Control: GitHub, GitHub Actions, Git and pull-request workflows
Observability & Evaluation: Tracing, evaluation harnesses, LLM observability, logging and monitoring
ITSM & Agile Tooling: ServiceNow, Jira
Scripting & Languages: Python, PowerShell
CERTIFICATION
Cloud or AI/ML certification (Azure AI Engineer, AWS Machine Learning, or Databricks) (Preferred)
WORK LOCATION:
Hybrid schedule in Taylor, TX 2 days per week.
#LI-DN
ERCOT is firmly committed to equal employment for all qualified persons without regard to race, sex, medical condition, religion, age, creed, national origin, citizenship status, marital status, sexual orientation, physical or mental disability, ancestry, veteran status, genetic information or any other protected category under federal, state or local law.
Expected Salary Range:
$145,000 - $200,000