Reports to: EVP, Technology & Innovation Location: On-site in Columbia or Kansas City, Missouri preferred.
ESS Companies is a heavy civil construction holding company with six operating subsidiaries. We build roads, bridges, and infrastructure — and increasingly, we build the software and AI systems that make that work faster, safer, and more profitable.
AI and automation are changing how work gets done, and we intend to be deliberate about it rather than swept along by it. This role owns that: finding where AI genuinely helps the business and then making it happen. It is not a research role and it is not a slideware role. It's a builder's role — you'll investigate problems, design solutions, and ship them, sometimes as code, sometimes as a configured tool, sometimes as a vendor you talked us into buying.
You'll be early. You will not be employee #40 on an established AI team — you'll be the start of one. As the work matures and the function grows, this role has a clear path to leading the team it seeds. You'd work directly with our EVP of Technology & Innovation to shape what that team becomes and to help build it out. We want someone who can be a strong individual contributor now and grow into that leadership as the function scales — not someone who needs a team underneath them to be effective on day one.
If "define the playbook as you go" sounds like too much ambiguity, this isn't the job. If it sounds like the fun part, keep reading.
This role spans four kinds of work. You'll move between them constantly.
1. Get more out of the tools we already have. We run Microsoft Copilot, Claude, and ChatGPT across the enterprise. Most people use a fraction of what these can do. You'll build agents, projects, and workflows on top of them, write the prompts and guardrails that make them reliable, and help ~3,000 professional staff actually adopt them.
2. Connect AI to our data. Our analytical source of truth is an enterprise data warehouse, fed from our ERP and other source systems. Our documents live in cloud file-sharing and collaboration platforms. You'll wire AI tools into these sources — via APIs, connectors, and retrieval pipelines — and expose them as agents and projects people can use without a data engineering degree.
3. Evaluate and drive commercial (COTS) solutions. Not everything should be built. You'll assess vendor AI products, run honest build-vs-buy analyses, and when something makes sense, drive the implementation across the relevant subsidiaries. This requires the credibility to sit in an operations meeting and be believed.
4. Build custom automation and applications. When there's no good tool to buy, you build it. Our stack is Google Cloud Platform (Cloud Run, Cloud SQL/PostgreSQL, Pub/Sub, Cloud Scheduler, Secret Manager), BigQuery and dbt for data, and a CI/CD-driven monorepo. AI shows up two ways here: sometimes it's in the product (extraction, classification, agents), sometimes it's just how you build faster (we use Claude Code heavily), and often both. You're expected to be fluent in agentic development workflows, not just aware of them.
- You can actually build. You write production code (Python and SQL at minimum), deploy it, and maintain it. You're comfortable in the cloud — GCP preferred, but we'll take strong AWS/Azure if you can switch.
- You're fluent with modern AI tooling: LLM APIs, prompt design, retrieval/RAG patterns, agent frameworks, and AI-assisted development (Claude Code, Copilot, or equivalent). You understand where these tools are reliable and, more importantly, where they aren't.
- You can work with data: SQL is second nature, you understand warehouses and ELT, and you can model and query messy real-world business data.
- You can translate. You can talk to a project manager about job costing, then go write the code, then explain the result to a CFO. The translation is half the job.
- You have good judgment about build vs. buy, and you're not religious about either.
- You operate with minimal direction. You can take a vague business problem, scope it, and come back with something that works.
- Construction, engineering, or other operations-heavy industry experience.
- Experience integrating with enterprise systems such as ERP (Viewpoint Vista or similar) and HCM platforms (Workday).
- dbt, BigQuery, and modern data stack experience.
- Experience standing up internal-facing applications with enterprise auth (Entra ID / OIDC).
- A track record of getting non-technical people to adopt new tools — adoption is harder than building.
The person who does well in this role is pragmatic over flashy, ships over theorizes, and would rather deliver a credible $100K improvement than promise an unsupportable $1M one. You're comfortable being early, you don't need a playbook handed to you, and you can hold your own with operators who are skeptical of anything with "AI" in the name.
Equal Opportunity Employer/Protected Veterans/Individuals with Disabilities
The contractor will not discharge or in any other manner discriminate against employees or applicants because they have inquired about, discussed, or disclosed their own pay or the pay of another employee or applicant. However, employees who have access to the compensation information of other employees or applicants as a part of their essential job functions cannot disclose the pay of other employees or applicants to individuals who do not otherwise have access to compensation information, unless the disclosure is (a) in response to a formal complaint or charge, (b) in furtherance of an investigation, proceeding, hearing, or action, including an investigation conducted by the employer, or (c) consistent with the contractor’s legal duty to furnish information. 41 CFR 60-1.35(c)
