Loading...
Loading...
Billings sits at the crossroads of the Northern Rockies and the High Plains, and its economy reflects that blend: oil refining along the Yellowstone River, regional healthcare anchored by Billings Clinic and St. Vincent Healthcare, agribusiness moving cattle and sugar beets, and a logistics network that touches three states from a single distribution belt. AI work here is rarely about chasing the next foundation model. It is about reducing unplanned downtime at a Phillips 66 or ExxonMobil refinery, tightening claims and imaging workflows at a regional health system, or pulling sensor data off harvest equipment moving through the Heights. The professionals who do this work tend to know the operators by name and have spent time on the catwalks, in the clinics, or in the grain elevators they are trying to instrument.
Billings does not market itself as a tech hub, and that is part of the appeal. The professionals working in machine learning here came up through Montana State University Billings, MSU Bozeman computer science transfers, or remote roles tied to Denver, Salt Lake, and the Twin Cities. What you see locally is a small but tight community of consultants and in-house engineers concentrated downtown along North 27th Street, around the Rocker Six Carbon coworking space, and in the office parks off Shiloh Road on the West End. The regional economy is dominated by Billings Clinic, St. Vincent Healthcare, ExxonMobil's Billings Refinery, the Phillips 66 refinery, CHS Inc., and First Interstate BancSystem, which is headquartered here. That mix produces a specific flavor of AI work: predictive maintenance on aging refining infrastructure, claims and prior-authorization automation in healthcare, credit and fraud modeling for a multi-state community bank, and demand forecasting for agricultural cooperatives. The Big Sky Economic Development Authority and the Rocky Mountain College computer science program both feed talent and partnerships into that ecosystem, though most senior engineers came into Billings from somewhere else and chose to stay.
Energy and refining is the most active sector. The ExxonMobil Billings Refinery and Phillips 66 Billings Refinery both run continuous-process operations where unplanned shutdowns cost six figures per hour, and ML on vibration, temperature, and flow sensor data is finally moving from pilot to production. Engineers here work with historian data from systems like OSIsoft PI, often alongside reliability engineers who have decades of plant history but limited Python experience. Healthcare is the second concentration. Billings Clinic operates as a tertiary referral center for a region larger than most U.S. states, and its imaging volume, telehealth footprint, and electronic record system create real opportunities for clinical NLP, radiology triage, and patient risk stratification. St. Vincent Healthcare and Intermountain affiliates are exploring similar territory, with cautious adoption shaped by HIPAA, rural broadband limits, and a patient population that skews older and more dispersed than national benchmarks. Agriculture, banking, and logistics round out the picture. CHS Inc. and regional cooperatives apply ML to grain quality, yield forecasting, and rail logistics. First Interstate Bank uses analytics for credit decisioning and small-business lending across its multi-state footprint. Distribution operations near the airport and along I-90 use route optimization and warehouse forecasting models, often built by small consulting shops rather than large vendors.
The Billings AI talent pool is small, and that changes how you should approach hiring. Expect to find a mix of full-time engineers at Billings Clinic, the refineries, and First Interstate, plus a handful of independent consultants and small firms doing project work for agribusiness and mid-market clients. Pure research backgrounds are rare. What you will find more often is a senior engineer who has spent five or more years inside one of the dominant industries and can speak the language of the operators they support. If you are a hiring manager, ask candidates to walk you through a project where the model was the easy part. The hard work in Billings tends to be data access, sensor reliability, change management with field teams, and coordinating with vendors who built the original SCADA or EHR systems. A consultant who has shipped models to a refinery control room or a clinical workflow will describe those constraints in detail. One who has only worked on academic datasets will not. Typical engagement structures lean toward fixed-scope projects rather than open-ended retainers, partly because budgets are smaller than coastal markets and partly because Billings clients want clear outcomes. Hourly rates for senior independent AI consultants generally land between $150 and $225, with full-time senior ML engineering salaries running $130K to $170K depending on industry. Refining and banking pay at the top of that range; nonprofits and smaller agribusiness clients sit at the bottom.
For a team of two to four people, yes, particularly if you are open to a mix of full-time and contract talent. Billings has senior ML engineers placed at Billings Clinic, the two refineries, and First Interstate Bank, plus a small consulting community downtown. Beyond that, most companies blend local hires with remote contributors based in Denver, Boise, or the Twin Cities. Building a team of ten or more entirely from local talent is difficult and usually unnecessary; a hub-and-spoke model with a senior local lead and remote engineers is more realistic and matches how most successful Billings AI groups have actually scaled.
Expect a heavier dose of legacy systems than you would in a coastal market. Refineries run historian platforms like OSIsoft PI alongside older SCADA stacks. Healthcare clients are usually on Epic or Cerner with limited data warehouse maturity. Banks frequently use mainframe or older core banking systems with batch-oriented data flows. Agribusiness clients may have sensor data scattered across vendor portals. The first phase of most engagements in Billings is data plumbing, not modeling. Consultants who can stand up modern ELT, build a usable warehouse, and earn trust with the existing IT team often deliver more value than a pure modeling specialist.
The community is informal but real. Rocker Six Carbon Coworking on Montana Avenue hosts occasional tech meetups, and the Big Sky Economic Development Authority runs periodic innovation events that draw the local data and AI crowd. MSU Billings and Rocky Mountain College both run guest speaker series in computer science. Beyond that, most Billings practitioners participate in regional virtual communities centered on Bozeman, Missoula, and Denver, where the meetup density is higher. There is no large annual AI conference in Billings, but quarterly informal gatherings and refinery-adjacent reliability conferences regularly feature ML content.
Bozeman has the strongest pure tech and startup density in Montana, driven by MSU's computer science program, photonics companies, and a steady inflow of remote workers. Missoula has a smaller but vibrant scene tied to the University of Montana and a creative tech community. Billings differs by being more industrial: the work is in refining, healthcare, banking, and agribusiness rather than software product companies. If you are looking for a research-oriented ML scientist, Bozeman is a better starting point. If you need someone who can ship production AI inside a regulated, asset-heavy environment, Billings has more relevant operators.
Most Billings engagements are paced by the customer's operational calendar more than the consultant's roadmap. Refinery projects often align to turnaround windows, which means a discovery phase in spring, model development through summer, and integration into control workflows during fall outages. Healthcare projects move at the speed of compliance review, typically four to six months from kickoff to first production deployment. Banking and agribusiness projects are quicker, often three to four months for an initial production model. Plan for a discovery phase of four to six weeks regardless of vertical; skipping that step is the single most common reason projects stall here.
Reach businesses across Billings's 119,116 residents.