The Expert Opinion provides a platform on NaturalRefrigerants.com for experts to share their views on technology, market trends, policy developments and more. Contributors are not paid, and all opinions are their own. NaturalRefrigerants.com maintains full editorial control over each submission. If you’re interested in contributing to The Expert Opinion, please fill out this form.
This Expert Opinion column has been written by Amrit Robbins, CEO and Co-founder of Axiom Cloud, whose AI-powered refrigeration management software is used by grocery stores and cold storage facilities to enable predictive maintenance and early leak detection. Here, Robbins shares his thoughts on how a new approach to recruiting and AI-powered tech can help accelerate the transition to natural refrigerants in the U.S.
The technician shortage: an untapped opportunity
The technician shortage hits natural refrigerants much harder than HFCs. With traditional systems, you’ve got a large workforce with decades of experience, but natural refrigerants require completely new and specialized training and certifications.
Industry data shows we’re facing a shortage of about 110,000 HVAC&R technicians nationwide, with 25,000 leaving the workforce annually. This creates serious bottlenecks such as longer response times and higher service costs. In rural areas where the shortage is especially acute, you might not find a qualified natural refrigerant technician at all.
That’s a major reason why we’re looking at a 20 to 30 year transition timeline. The shortage isn’t just slowing adoption. It’s a fundamental constraint that makes companies hesitant to deploy systems they’re not confident they can maintain.
That said, we’re sitting on a massive untapped opportunity to attract young people to the refrigeration technician field. You’re offering young people the chance to work with genuinely advanced technology, make meaningful climate impact with every service call and enter a field desperate for workers with virtually guaranteed employment. I definitely do not predict a future where AI replaces refrigeration technicians. It’s just not going to happen anytime soon, or in the foreseeable future.
Yet the industry still markets this career like it’s 1985. What we need is a complete rebrand, from “person who fixes leaks in grocery store back rooms” to “sustainability technician working on next-generation cooling systems.” The climate angle is particularly powerful because this isn’t greenwashing.
When you can tell someone their work directly prevents emissions equivalent to taking thousands of cars off the road, that resonates. We need to partner with technical schools differently and show young people that this is a tech career with purpose. The workforce crisis won’t solve itself with better pay alone. We need to make the work itself attractive to a generation that cares deeply about climate and technology.
AI-powered predictive maintenance: key to reducing the natural refrigerant learning curve
Natural refrigerant systems are intimidating for newer technicians. They’re often called to catastrophic failures without adequate context, leading to multiple site visits to diagnose the same issue. Advanced diagnostics change that by helping them to prepare for each service call. This technology essentially provides the diagnostic expertise of a 20-year veteran who doesn’t exist in your market.
An AI-trained system can tell them exactly what’s wrong, where the problem is located, what parts they’ll need, whether it’s CO2-specific components or ammonia safety equipment. Instead of a five-year learning curve requiring apprenticeship under a master ammonia technician who may not be available, you can have someone productive within months.
For CO2 (R744) or ammonia (R717) systems where there aren’t enough experienced mentors to go around, this technology becomes a critical tool that can enable less-experienced technicians to succeed much more frequently. When a truly unexpected emergency occurs, the right predictive maintenance technology can help an inexperienced technician succeed in a pinch by guiding them every step of the way.
The key is real-world training data from actual installations, not just engineering manuals or “best practices.” The best systems learn from thousands of site-years of operational data across different natural refrigerant types, climate zones, equipment manufacturers, system configurations, control algorithms, store sizes and equipment ages.
Because solutions like Axiom Cloud’s analyze real-time data from existing refrigeration controllers combined with each facility’s operating history, they learn what “normal” looks like for that specific system type and configuration.
The best AI-powered predictive maintenance solutions are also trained using datasets of tens of thousands of historical problems or anomalies that are labeled by refrigeration experts. For each problem, the refrigeration specialist labels the start time, the end time and the root cause. That is how the AI learns to recognize the leading indicators of future problems before they occur.
It’s about understanding how transcritical CO2 systems actually perform across different ambient temperatures and load profiles, or recognizing abnormal ammonia purge cycle anomalies that may seem totally normal to inexperienced operators. The result is system-specific diagnostic insights rather than generic alerts that waste technician time.
The business case for end users
The economics become even more compelling for natural refrigerants precisely because of the technician shortage. Consider the risk equation: When a CO2 or ammonia system goes down and there are only a handful of qualified technicians within 200 miles, you’re looking at potentially days of downtime, not hours.
Product loss alone can hit six figures at a single location. Service rates for specialized natural refrigerant technicians run 30-50% higher than standard HVAC rates, emergency calls cost even more and you’re often paying travel time from distant markets.
Predictive maintenance mitigates that risk by catching problems before they become emergencies, allowing operators to schedule service when qualified technicians are available rather than scrambling during crises. The avoided cost of a single prevented failure can justify significant investment in monitoring technology.
Beyond direct cost savings, there’s the operational continuity value, being able to deploy natural refrigerants confidently knowing you have systems in place to keep them running when specialized help is scarce. For retailers serious about transitioning to natural refrigerants, predictive maintenance isn’t optional anymore. It’s the enabling technology that makes the transition operationally feasible.
