Mainframe workload modernization startup Mechanical Orchard Inc. said it’s hoping to gain more enterprise traction today after closing on a $24 million early-stage funding round led by Emergence Capital.

The Series A round, which values the startup at $95 million, also saw participation from Industry Ventures, Spider Capital, Bloomberg Beta, Cendana Capital and Webb Investment Network.

The startup is pioneering an artificial intelligence-based approach to help companies lift legacy workloads from their old, on-premises mainframes, into a cloud environment. The approach involves reverse-engineering many of those workloads, with an aim to reduce significantly the risk to companies through its superior visibility and control.

What’s more, the migrated systems can remain online throughout the entire process, meaning customers benefit from faster time to market. By moving mainframe workloads to the cloud, customers can take advantage of improved commercial and development capabilities, the company says.

Mechanical Orchard is led by an impressive team of technology industry veterans, headed up by its co-founder and Chief Executive Rob Mee. He made his name as the founder of Pivotal Labs Inc., the agile software developer that was ultimately acquired by VMware Inc. for $2.7 billion in 2019.

Another big name is Chief Operating Officer Matthew Work, who co-founded the startup alongside Mee and has previously held leading roles at Amazon.com Inc. and Cognizant Inc. The startup’s chief scientist is Kent Beck, who helped create the Extreme Programming agile software development methodology.

Mechanical Orchard takes advantage of the popularity of open-source software and AI models to transform the process of legacy workload modernization. Its platform relies on AI to make sense of the billions of lines of archaic code found in mainframe applications, then speeds up the process for engineers who have to rewrite that code and make it cleaner for its new cloud environment.

Speaking to Forbes, Mee said the arrival of generative AI brings forth the possibility of a “second major productivity revolution in software development.” He added that this is happening outside of the mainframe. “Folks with a fossilized IT substrate are going to miss out,” he said.

While Mee’s old company, Pivotal Software, was also focused on application modernization, but it failed to crack the decades-old mainframe industry, which remains an extremely big business. A 2021 report from IBM Corp. notes that 45 of the world’s 50 biggest banks, and seven of the 10 largest retailers on the planet still rely on mainframes for key business functions such as payroll and customer purchases.

Mee said he was inspired to launch Mechanical Orchard after helping a number of state and local governments move their existing systems from mainframes to the cloud during the COVID-19 pandemic. Under pressure, those organizations were looking for a way to migrate quickly to the cloud, so they could track COVID cases more quickly and provide assistance to sick citizens. Mee’s solution was to reverse-engineer and rebuild those systems as cloud-based applications, and the technology he developed forms the basis of his new startup.

The process of building a digital twin of existing on-premises systems begins with identifying a system or job on a mainframe and assessing its code for possible improvements in a test environment. Then, the startup uses its AI models to identify the essential functionality of that system, exposing many of the potential traps in its legacy codebase. Once those pitfalls are identified, it will then use a combination of generative AI and human expertise to create fresh code for that application, enabling it to be moved to the cloud.

Mechanical Orchard says its modernization process is faster and more cost-effective than rebuilding an entire system from scratch using an army of consultants. It also produces code that’s more coherent than entirely automated approaches. “No line of code gets written that hasn’t been tested,” Mee said.

Analysts were divided on the prospects for Mechanical Orchard and the potential impact it will have on the thousands of enterprises that still rely on mainframes for their most mission-critical workloads. Holger Mueller of Constellation Research Inc. said generative AI has already shown its revolutionary capabilities in several ways, and one of the most impressive is its coding abilities. “It’s no surprise to see generative AI being applied to code portability, or what essentially amounts to re-writing it,” Mueller said. “This is what Mechanical Orchard is doing, and the results so far are very impressive. But the startup will have to show it is capable of migrating larger workloads if it’s too convince enterprises of its abilities.”

Charles King of Pund-IT Inc. said mainframe modernization has, until now, generally come in two flavors, with the first being hardware-based solutions that amount to replatforming initiatives. “This is where the aim is to move IBM mainframe customers to alternative, mainly x86-based servers,” King explained. “Those efforts mainly work for businesses that are no longer growing their mainframe-based applications.”

Then there are software-based solutions to mainframe modernization that are more focused on cloud and AI. “The question for enterprises is whether or not these solutions are superior to IBM’s existing alternative, Watsonx Code Assistant for Z, which the company announced in December,” King said. “I have great respect for Rob Mee and what he did with Pivotal, but if companies are being asked to choose between a promising startup or the longtime leader in mainframe systems and innovation, namely IBM, my money is on the latter.”

Although it faces a tough competitor in IBM, the startup has made good progress so far. Founded just over 16 months ago, Mechanical Orchard was already raking in more than $10 million in annual revenue at the end of last year, and is also profitable, thanks to strong demand from enterprise customers such as the logistics giant Omni Logistics LLC.

“The prospect of modernizing our IT infrastructure often kept me up at night,” said James Ryan, Omni’s vice president of software engineering. “Mechanical Orchard’s team made quick work of dissecting our legacy systems, core business processes, and charted a course for our future technology. Their disciplined approach to solving the hard problems in software engineering is unparalleled in the industry.”

The startup said the funds from today’s round will go toward continuing its platform research and development, and building a comprehensive global sales and marketing plan. The company will also expand its team of more than 50 employees.

Emergence Capital General Partner Jake Saper said AI has unlocked an entirely new chapter in the story of IT transformation. “Mechanical Orchard is now the company at the forefront,” he said. “It’s applying AI meaningfully to solve what was considered an impossible challenge.”

Image: Mechanical Orchard