AI-Powered Salesforce Testing, Shocking Agile Failure Rates, and More TGNS124

[00:00:00] Joe Colantonio Have you seen the study that shows 268% higher failure rates for agile software projects? What automation tool just announced a new AI-driven solution for Salesforce test automation? And how do you scale with AI? Find out in this episode of the Test Guild News show for the week of June 9th. Grab you favorite cup of coffee or tea and let's do this.

[00:00:23] Joe Colantonio I get us all the time. Hey Joe, how do I automate Salesforce? Well, here's one new solution you definitely should check out. And it's all about Testsigma's new Salesforce Test Automation. And what's great about the solution is that it's built on top of Testsigma's gen AI powered low code platform, and this new tool aims to simplify and expedite Salesforce testing, enabling users to automate test up to ten times faster than traditional methods, according to Testsigma. This platform utilizes a natural language programing engine, allowing tests to be written in plain English. This really helps to simplify and ensure that those with limited technical knowledge, especially those with business knowledge, who have to deal with Salesforce apps all the time, can now feel comfortable and confident and using these tools to help them with testing. Additionally, Testsigma leverages Salesforce metadata and APIs to create robust and adaptive test scripts, reducing maintenance and improving reliability. As you can see, this new Salesforce Test automation tool offers a bunch of benefits for testers, so I highly recommend you check it out using the link Testguild.me/testsigma and you can find that link down below.

[00:01:34] Joe Colantonio So I'm dying to know what you think about this new study that reveals higher failure rates for agile software projects. And it is mind blowing. How much so? Well, let's check it out. This recent study reveals that software projects using agile methodologies are 268% more likely to fail compared to those that don't. So this research analyzed responses from 600 software engineers in UK and the US, and it found that 65% of agile projects failed to deliver on time and within budgets, whereas only 10% of projects using a new impact engineering methodology failed. The study also highlights the importance of robust requirements engineering and psychological safety for successful project outcomes. Agile practices such as working without a clear specification and making significant late stages changes were linked to a lot higher failure rates, but in contrast, clear project requirements and the ability to address problems promptly improved success rates significantly. And according to this article, this approach reduces the risk of defects in late stage changes, ensuring more stable and predictable project outcomes. Is it true? I don't know. Check it out for yourself and let me know your thoughts in those comments down below.

[00:02:48] Joe Colantonio So the king of AI, Tariq King just released a new blogpost. What's it about. So this article posted on epam focuses on the adoption of multi-modal test strategies to ensure the quality and reliability of AI systems. And Tariq goes over how multimodal AI, as shown by models like GPT four turbo, involves the use of various forms of data like images, video, text, speech, audio, and it makes predictions and draws insights from these. And this approach is providing to be a standard for many organizations. And Tariq also goes of a shift left testing if you don't know, which emphasizes addressing quality earlier in the software development lifecycle to minimize late defects. And for AI systems, it is crucial to test prerelease, and conduct post-release production testing. This do approach outlined in this article ensures ongoing quality and performance in dynamic environments, and you can find out more about it in the link down below.

[00:03:48] Joe Colantonio Just saw this next article on LinkedIN and posted by David Burns. And he linked to how Firefox is just deprecated CDP support and is transitioning to WebDriver BiDi, which I think is a trend that a lot of companies and a go towards. The announcement goes over that starting with Firefox version 129 support for Chrome DevTools protocol CDP will be deprecated. And users are encouraged to migrate to the W3C WebDriver BiDi protocol, so it also goes over how the transition to WebDriver BiDi is part of Mozilla's efforts enhance cross-browser testing automation. WebDriver BiDi provides superior features compared to Firefox experimental CDP implementation, including improved Dom updates, network interception, and better logging capabilities. If you haven't already, you should definitely check out the benefits from the enhanced capabilities of WebDriver BiDi, because it's a thing that a lot of companies are going to be transitioning to, and you could definitely find out more about it in this article for a deeper dive.

[00:04:47] Joe Colantonio I also found another article on LinkedIN from Abhinaba about Git Labs test platform for teams to validate AI features. Vincy and Mark share insights for their approach to validating AI. The team uses a blend of automated and manual testing techniques to ensure the reliability performance of new AI functionalities. And this approach includes vigorous testing, protocols, continuous integration practices, and synthetic data to simulate real world scenarios through a strategic integration of AI into the testing process. The team is able to swiftly identify and resolve potential issues. This not only enhances the overall quality and efficiency of the software development lifecycle, but also underscores the significance of thorough validation processes and deploying AI-driven features. Just another great real world scenario of how a team is testing AI that you might be able to learn from it. You can find out more about it in the links down below.

[00:05:45] Joe Colantonio I also found a new announcement about an AI code generation tool, and this is by Mistral AI that has introduced Codestrail, which is a new AI driven tool to assist developers and testers in code generation task. And it goes on how Codestrail supports over 80 programing language, including all the popular ones like Python, Java, and C++ and more niche ones like Swift and Fortran. And the model aims to streamlining not only coding by completing functions, but also helping to write test and filling in partial code, reducing errors and saving time. And it's currently available for download on Hugging Face under the Mistral AI non-production license for research and testing, and it also integrates with popular development environments like VSCode and JetBrains.

[00:06:29] Joe Colantonio And if you have anything to do with Appium and Flutter, you're going to be really excited about this next announcement. So Sai just announced a new open source driver to enhance Appium and Flutter automation. And the new open source tool is called Appium Flutter Integration Driver. And this was designed to simplify the automation of applications built for Appium and Flutter. And this tool integrates with Appium's powerful automation feature with Flutter's modern app built in capabilities. Leveraging the flutter integration test framework. It uses Appium. This is great news because testers can now easily automate testing for Flutter applications reducing manual test time. I think it's a must check out driver for all mobile testers that you should be more familiar with. And once again, you can find that in the links down below.

[00:07:16] Joe Colantonio If you have to deal with enterprise grade pipelines, I have another announcement that you might be excited about. It's how snowflake has announced new tools and innovations to enhance developers capabilities, working with enterprise grade developer pipelines, models, and AI powered applications, and announce it at its annual user conference. These updates include snowflake notebooks, Snowflake Trail, and various DevOps tools, all integrated within Snowflake's unified platform, and the aim is to simplify development processes and accelerate development of AI applications. And Jeff Holland, the head of application and development platforms at Snowflake, talked about how this really is poised to redefine the boundaries of what developers can accomplish with AI data cloud.

[00:07:58] Joe Colantonio And I also found out this next announcement about AppDynamics on LinkedIn from Rebecca and Rebecca, posted a link to how AppDynamics has introduced a Log Observer Connect, which they new integration with Splunk that enhances log base troubleshooting with Cisco's full stack observability solution. This integration allows users to centralize log collection in Splunk, streamlining the process of identifying resolving application performance issues, and the unified platform aims to reduce operational costs and improve root cause analysis. And the collaboration leverages Splunk Log Analytics to provide more detailed insights. Its application performance issues improve in the meantime to resolution.

[00:08:38] Joe Colantonio And last up is an article all about scaling businesses with AI in the hybrid cloud. This was posted by Arvind, IBM's CEO, which is now a significant advancement in AI and hybrid cloud solutions at IBM think 2024 event. And it goes over how the company is open sourcing its powerful AI models. The Granite family on HuggingFace in GitHub under Apache two license, and these models are designed to push the boundaries of AI capabilities. And it also goes on how IBM introduced a suite of new Watson X assistants powered by granite to enhance productivity and streamline operations, and this includes tools for coding, deploying custom AI assistants, and improving interactions with mainframe systems.

[00:09:20] All right, for links of everything of value we covered in this news episode. Head on over to the links in that comment down below. So that's it for this episode of The Test Guild News Show. I'm Joe, my mission is to help you succeed in creating end-to-end full-stack pipeline automation awesomeness. As always, test everything and keep the good. Cheers.