# Joe Name **AI-First Integration & Infrastructure** [email protected] • linkedin.com/in/joename --- I build AI-enabled systems that solve real problems in government, legal, and infrastructure contexts. My approach is AI-first—not as a buzzword, but as a practical stance: every workflow I touch gets evaluated for where AI integration creates genuine leverage, and where it doesn't. Decades of infrastructure experience—Linux administration, CI/CD, cloud services, monolithic and microservice architectures—give me the foundation to build integrations that actually hold up in production. I've managed the systems; now I make them smarter. ## Current Focus **Custom RAG Systems** Designing and building retrieval-augmented generation pipelines for legal code and government data. First iteration handled plain-language querying of Indiana statutes; the rebuild addresses structural and retrieval problems found in production, with expansion to additional states. **AI Integration Across Workloads** Evaluating and embedding AI tooling into varied workflows—development, document processing, legal research, stakeholder communication. I work with Goland, Cursor, and Claude Code daily and build custom integrations where off-the-shelf tools fall short. **Practical AI Safety** Building and testing workflows that probe how AI systems behave under real conditions. Focused on misuse potential, incentive alignment, and the gap between vendor claims and actual capability. ## Perspective I have a strong bias toward substance over presentation. AI is the most significant tooling shift in decades, but that makes clear thinking about it more important, not less. I focus on what these systems actually do—where they create real value, where they introduce risk, and where the gap between marketing and reality needs to be closed before someone gets hurt. _My experience spans decades. When the machines take over, I won't be saved—I know where too many digital bodies are buried._