Director Prompt Engineer Jobs (NOW HIRING)

About H&W

H&W Electrical Corporation is an industrial and commercial electrical contractor serving South Carolina, North Carolina, Tennessee, Georgia, and Florida. Headquartered in Greenville, SC. People come first here — it shapes how we hire, train, and promote. We invest in our team through H&W University, our internal development program, and we expect everyone, office and field, to grow.

The Role

The Prompt Engineer owns the practical AI capability that supports estimating, project management, engineering, procurement, business development, and executive leadership. You will research, build, validate, and deploy prompts, AI applications, and agents that take friction out of our day-to-day workflows. You will track AI/ML developments in construction and convert them into disciplined adoption decisions. You will contribute AI-literacy instruction to H&W University.

This is a build-and-ship role. Research without a path to deployment is not what we need. The work touches nearly every department, so the seat is wrong for someone who prefers to work in isolation.

What You'll Do

• Research, compose, test, and refine production prompts; package successful patterns into a version-controlled prompt library

• Develop and maintain AI-powered applications that address specific operational needs across the company

• Build and optimize AI agents for specific roles across the company (estimators, PMs, procurement, BD, field leadership)

• Develop AI-assisted document analysis tools that support estimating, project management, and engineering review workflows

• Develop and operate internal intelligence tooling that supports business development and executive decision-making

• Evaluate AI/ML platforms, run time-boxed proofs of concept, and deliver written recommendations with ROI analysis

• Stand up local LLM infrastructure where it makes sense for cost, latency, or data sensitivity; decide when to fine-tune vs. prompt engineer vs. retrieve

• Design and maintain the data layer behind your tools — SQL, SQLite, PostgreSQL, vector stores, flat files — sized to the problem

• Develop and integrate with MCP (Model Context Protocol) servers connecting LLMs to internal data sources and business applications

• Attend AI conferences and vendor trainings; bring the field's best ideas back into H&W

• Produce documentation — prompt specifications, validation results, READMEs, user guides — for every production deliverable

• Co-deliver AI-literacy modules in H&W University and partner with our Business Applications Specialist on AI-integrated automations

Required Qualifications

• Fluent in Python — you can read, write, and ship production-grade code without supervision

• Working knowledge of at least two commercial LLM platforms (Anthropic Claude, OpenAI, Google Gemini, Meta Llama, Mistral, or comparable) and the differences between them

• Direct experience designing, validating, and deploying prompts against representative datasets — not hobbyist usage

• Familiarity with MCP servers, function calling, and tool-use patterns

• Database architecture experience across the spectrum: relational (SQL, SQLite, PostgreSQL), vector databases (Pinecone, Chroma, Weaviate, FAISS, or pgvector), and flat-file (CSV, JSON, Parquet)

• Experience fine-tuning open-source language models and running them locally (Ollama, LM Studio, llama.cpp, vLLM, or similar)

• Working knowledge of retrieval-augmented generation (RAG), embedding models, and at least one agent framework (LangChain/LangGraph, CrewAI, AutoGen, or equivalent)

• Git version-control discipline — meaningful commits, branch hygiene, and README-grade documentation

• Strong written communication. You can produce a prompt specification, a validation report, and a one-page executive summary for the same project

• Comfort with ambiguity — you can take a vague problem from a department head and turn it into a scoped, shippable deliverable

Preferred Qualifications

• Background in a role where you had to read people in real time and adapt on your feet — teaching, hospitality, service, healthcare, retail management, or similar. Our internal customers are coworkers, not tickets, and the people skills carry over

• Experience training, teaching, or onboarding others

• Construction, manufacturing, or other industrial/operational environment experience

• A portfolio of shipped side projects, OSS contributions, or applied AI builds outside of work — show us what you've made

• JavaScript/TypeScript for front-end AI interfaces; Docker for packaging

• Experience with API development (FastAPI, Flask) and document parsing libraries (PyMuPDF, pdfplumber, Unstructured)

Bonus (Not Required)

• Blockchain and cryptographic primitives, smart contracts, or Web3 application development

• Experience deploying AI workloads on decentralized compute (Internet Computer / ICP, Akash, Bittensor, or similar)

• Cybersecurity background — threat modeling, secure coding, secrets management, supply-chain hygiene

How to Apply

Submit your resume and a short note about a project — yours or your team's — that you would point to as your best work. A working demo, a repo link, or a write-up is welcome in place of a traditional cover letter.