You should take a look at my company. Heirloom Computing. Heirloom.cc We have migrated many mainframe application and millions of lines of cobol and pl1 into Java and deployed it into production on prem and into the cloud.
It's a demo but the code that is transformed is trivial in the extreme. The whole approach of selecting small pieces is just not scalable. One small project i worked in was 22 million lines of cobol code, spread over 4000+ programs imagine having to select bits of code out of that.
Heirloom Computing, I am CTO, have transpilers for cobol and pl/1. code transpilation is the easy part. Create a Grammer for the source language, generate the ast, transform the ast to the target language ast and generate the code.
Now make that generated code run and behave the same way as on the mainframe and give it all the subsystems that exist on the mainframe for transaction, file handling, security etc. This is the part that is difficult. Fortunately it's already solved and in production with our clients.
So transpilers do work and are in production but our biggest competitor is inertia.
Heirloom computing where I am cto does this using transpilers with 100% automated transpilation. Using LLMs for an entirely deterministic domain borders on the insane. This is just marketing bs but we get asked about it and what our plan is to counter it all the time. Explaining that using Gen-ai and LLMs for what is a well understood compiler/transpiler problem that is already solved just seems to be too difficult for some people to understand.
Or any of the companies that can run mainframe COBOL and provide all the ecosystem. heirloom computing is of course the best (i am cto) but Micro Focus, lzlabs etc all enable you to run off mainframe with a full ecosystem.
Do not know about our competitors but you can run our product for free on a raspberry pi if you just want to play around.
well they still run on the mainframe, many of which can sit under your desk nowadays like a standard tower.
but there is a whole industry about replatforming and refactoring these apps to modern infrastructure and software stacks.
its not so much COBOL, its the mainframe environment that is brilliant for critical stuff. My company migrates mainframe applications to the cloud so we find all kinds of reasons for clients to move, cost/skills/scalability etc but the mainframe is a highly optimized platform for running mission critical applications. Its a different model 1 platform 99.9999% uptime vs the distributed cloud approach where you can have failures on any particular node. A lot of other stuff on the public cloud replicates and is inspired by the mainframe.
well i spend all my time moving people and applications off the mainframe so probably not the best person to ask, however a quick google of mainframe jobs showed up 100's (for experienced people). I have heard some large companies are instituting their own training programs for hew hires but i have no details.
In a word no. There is a place for LLM in mainframe migrations but it is not in source code transpilation.
COBOL (or PL/1) to any other language is a deterministic problem, imagine transpiling millions of line of code and every time you run the compiler you get slightly different output...
Anyway transpiling from one language to another ie COBOL to java is almost the easy bit, the hard bit is making the behavior and execution correct, especially when there are multiple OS and product services that don't exist outside the mainframe.
disclosure I am CTO of heirloom computing, we migrate mainframe applications to the cloud.