Job Description & Details
This is a serious platform-building role focused on scaling enterprise generative AI, RAG pipelines, and multi-agent workflows under strict governance. If you are tired of writing wrapper scripts and want to build production-grade AI infrastructure that actual business units rely on, this position in Plano offers a legitimate technical challenge.
What You'll Actually Be Doing
You will spend your days architecting and operationalizing the technical foundation for an internal AI accelerator. Your routine will involve partnering with data owners to ingest messy enterprise knowledge into governed knowledge bases, building complex retrieval pipelines, and designing multi-agent workflows that can reliably use tools, handle errors, and know when to stop and ask a human for help. You'll be working in two-week sprints, dealing head-on with hallucinations, tracking latency and cost-to-serve, and writing maintainable code that doesn't fall apart once the pilot phase ends.
The Core Tech Stack
You need absolute mastery over production Python and modern software engineering practices—we're talking rigorous testing, CI/CD, and robust API design. Beyond standard software engineering, you must have hands-on experience building RAG systems with vector or hybrid search and implementing multi-agent orchestration frameworks. The team heavily values observability, guardrails, and telemetry, so knowing how to instrument LLM evaluations and implement responsible AI controls is non-negotiable for this stack.
Interview Expectations
Expect the interview panel to drill down into your architecture decisions around context retrieval and agent state management. They will likely ask you to explain how you would design a retrieval pipeline that minimizes hallucinations and handles stale enterprise data, looking specifically for your thoughts on evaluation metrics and citation tracing. Another common line of questioning will test your ability to handle failure modes in multi-agent workflows, where they want to see how you define explicit tool contracts, validation checks, and human-in-the-loop review points when an agent inevitably goes off the rails.
Application Advice
Your resume needs to immediately prove you've built LLM systems that went to production, not just local Jupyter notebook prototypes. Explicitly highlight your experience with vector search, embeddings, RAG evaluation metrics, and agent-workflow platforms. Make sure to weave in keywords like "governance," "traceability," "CI/CD," and "Responsible AI" naturally within your experience bullet points to easily clear the ATS filters and signal to the engineering team that you understand enterprise-scale development.