One of the most tedious (but critical tasks) for software development teams is updating foundational software. It’s not new feature work, and it doesn’t feel like you’re moving the experience forward. As a result, this work is either dreaded or put off for more exciting work—or both. Amazon Q, our GenAI assistant for software development, is trying to bring some light to this heaviness. We have a new code transformation capability, and here’s what we found when we integrated it into our internal systems and applied it to our needed Java upgrades: - The average time to upgrade an application to Java 17 plummeted from what’s typically 50 developer-days to just a few hours. We estimate this has saved us the equivalent of 4,500 developer-years of work (yes, that number is crazy but, real). - In under six months, we've been able to upgrade more than 50% of our production Java systems to modernized Java versions at a fraction of the usual time and effort. And, our developers shipped 79% of the auto-generated code reviews without any additional changes. - The benefits go beyond how much effort we’ve saved developers. The upgrades have enhanced security and reduced infrastructure costs, providing an estimated $260M in annualized efficiency gains. This is a great example of how large-scale enterprises can gain significant efficiencies in foundational software hygiene work by leveraging Amazon Q. It’s been a game changer for us, and not only do our Amazon teams plan to use this transformation capability more, but our Q team plans to add more transformations for developers to leverage.
Team/Legal Counsel.. Of course, still no response back..).. To reiterate, my original claims/torts/grievances/proof/evidence were/are ironclad/strong enough to command/demand a very large/lucrative award/earnings amount on their own merits/rarity/validity/veracity/invincibility; your/Amazon’s delegatory/representative law firm/legal team has further compounded the issues/matters/accrued more valid/irrefutable claims/information and provided me with more proof/evidence to support/bolster my case by corresponding/communicating with me in a manner that has been littered with brazen prevarications, blatant disregard for your as well as Amazon.com, Inc.’s claims of conducting/resolving business/legal matters/issues in a objective/nondiscriminatory manner regardless of the party’s/parties’ involved race/gender (of course I am both black and a female/woman of color), as well as additional correspondences/communications and evidence/proof regarding/against your side of the spectrum/equation because of your/their attempts at Civil Rights violations in regards to my case/lawsuit/civil suit, etc.. Moreover, my legal counsel and I have also come into the possession of other forms/mediums of
Upgrading codebases is a notoriously time-consuming and complex process. AI tools like Google Gemini, Chat GPT, and now Amazon Q offer significant promise in streamlining this task for developers and IT organizations. Years ago, as an engineering lead at a large bank, I led the migration of critical web applications from the outdated Struts framework to Spring. This involved converting thousands of lines of code and conducting extensive regression testing. Vendor estimates for this project were both costly and time-consuming. To address these challenges, we developed a custom automation tool in Java to expedite the conversion process, saving our organization substantial time and resources. AI-powered tools have the potential to further enhance developer productivity, reduce costs, and improve the quality of codebase upgrades.
It is great achievement, would be interested in more details. The blog about Amazon Q writes that it leverages OpenRewrite. I assume that a huge part of the changes (80% to 90%) are done by deterministic Open Rewrite recipes. Is there a statistic about the amount of changes which were done with AI assistance? And maybe more details how LLM could support with the cases for which there are now OpenRewrite recipes? Birgitta Boeckeler Harald Aamot Jochen Schneider Peter Giese Jonathan Schneider
Most GenAI projects are getting killed at the proof of concept (PoC) stage itself and are unable to move beyond that- something Gartner has been saying for quite some time now. So this is good news for every employee/professional generally worried about GenAI. In the above mentioned situation/use case , where a software version update is required routinely- something that is mostly considered a boring job but also has high RoI, GenAI may actually make a positive difference, due to its significantly reduced timelines. It's very likely that even in this scenario, although the timelines & team size may reduce significantly, the project may still need to be supervised by a professional with a high degree of maturity of having done this successfully, previously. Doesn't look like a novice can handle this anyways.
Glass half full time... One of the reasons internal development staff is handcuffed from modernization is because all effort is spent on maintaining legacy systems and the company can't afford multiple development paths. It would be very easy to see this article and say, "lost jobs! oh no!" and companies should be wary of eliminating staff when one aspect of software architecture saves so much time and money. But here's the best part. The saved time/money should be used to modernize internal software and allow development teams to grow into new architectures. This is where the real value of GenAI comes in.
evidence/proof within the subsequent month(s) since this lawsuit/civil suit has been in progress, to further solidify/guarantee a definite and flawless victory, with detrimental irreparable/irredeemable harm to the reputation/legacy/brand of those whom my case/lawsuit/civil suit may concern at Amazon.com, Inc., Amazon.com, Inc. as a Corporation in and of itself as well as Amazon’s delegatory/representative Law Firm/Legal Team/Legal Counsel.. I will expect that those involved in my Amazon.com, Inc. lawsuit/civil suit, as well as their delegatory/representative Law Firm/Legal Team/Legal Counsel, will finally gain/regain their sense of adequate/reasonable comportment/composure and provide the adequate/appropriate/long overdue response(s) within three days of this post, which will be Tuesday August 27, 2024 (it has now been more than 2 months since my last correspondence/communications response from those at Amazon.com, Inc. whom this case/lawsuit/civil suit may concern as well as Amazon’s delegatory/representative Law Firm/Legal Team/Legal Counsel, when previous times I received a response within a week/several days..) or there CERTAINLY must be major intervention from the applicable/necessary outside organizations/agencies..
This is what AI alarmists don’t understand when they talk about AI eliminating all but a few human jobs. The massive, gargantuan, monotonous part of the work iceberg that is hidden below the surface and never got done. The work that should be done, that we’d love to get to, that would make so many other problems go away, if only we had tools that made the work cost-effective. Just look at the amount of COBOL written 50 years ago still running on mainframes. It’s not there because people like COBOL. It’s there because the people who understood the code retired 10 years ago. We desperately need these AI powered tools to tackle technical debt and make new development and experiments 10x cheaper, not so we can fire 90% of the staff, but so we can do 10x more work with the staff we have. Because there’s a vast amount of demand out there, and we’re only just getting started.
🚀 Wow, this is incredibly impressive! The impact Amazon Q is having on foundational software updates is nothing short of transformative. Reducing what used to take 50 developer-days down to just a few hours is a game changer, and the savings in developer-years and efficiency gains are mind-blowing. 💡 It’s clear that Amazon Web Services (AWS) Q is not only boosting productivity but also enhancing security 🔒 and cutting costs significantly 💰. Kudos to the team for pushing the boundaries and making such a remarkable difference in the software development process. 👏 This is innovation at its best!
hacker
4moThis is impressive and represents real value. That being said, as with any automation tool, there are going to be limitations that produce a long tail of outcomes that either can't be processed or get processed in a manner that produces a poor result. ex: An application can not be bumped from Java(x) to (y) one or more dependencies is EoL and is incompabitble with Java(y) and what to replace it with is non obvious. or - and the one I am more worried about: The automation makes a bad choice in how to update that breaks things in subtle ways. Perhaps it replaces a thread pool with a call to the common fork join pool in a Spring service and creates a bit of digital constipation that significantly nerfs thruput. The thing still compiles and runs - but now it runs poorly whereas before it did not and it is not obvious as to why until months later when you realize cloud spend has shot up because it takes 3x the instances to do the same work. Don't get me wrong - I see the value and am glad y'all are doing things like this. Just saying I will be following this closely to see where it struggles AND where it shines 🍻