Rafael E. Santos’ Post

In this Part 2 analysis of the AI software testing market, Rafael E. Santos explores various AI/ML approaches for automated test design, such as model authoring, user journey monitoring, and NLP requirements ingestion, highlighting their limitations and challenges. Rafael notes that these methods often rely on incomplete or outdated models. Some methods are incremental improvements that fail to address fundamental testing issues. Is there a path forward? Can the software testing industry successfully deal with the enormous challenge of significantly improving software testing through AI? You'll have to catch Part 3 to find out! #artificialintelligence #softwaretesting #qualityassurance #qa #sdlc #softwaredevelopment https://lnkd.in/eAh6B6XC

In 2024, AI 🧠 in software testing isn't just a buzzword—it's revolutionizing the field. But how exactly? Let's dive into together: https://lnkd.in/dTteXZcK #ai #testing #qa #aitesting #artificialintelligence

Mradul Kanugo, our Chief AI Officer, recently authored an insightful #Article on the website, "Artificial Intelligence Made Simple". It's special because it's one of the first pieces on #PromptTesting, and explores the detailed process of developing top-notch applications utilizing #LargeLanguageModels (LLMs). We commend Mradul for his diligent research and profound expertise in the domain. #Softude #AIExpert #LLM #AIApplicationDevelopment #TechInsights #AIInnovation #AI

Insightful article from Lorraine Steyn on using AI and human expertise to modernize older systems.

I think this issue will affect a lot more than Software Developers. It will ultimately create issues for most, if not all knowledge based work.  It is often said that the important part of knowledge work isn't the discipline itself but the critical thinking and problem solving skills, "Let the AI do the drudgery." One of the problems with that line of thinking is outlined in this article. Basically,  if one doesn't come up doing the work, one doesn't learn the domain well enough to be able to solve problems that are too complex for AI, to create a solution appropriate to the importance/scope of the problem,  or solve problems which are too new or too niche for an LLM to have been adequately trained on. I think there are a variety of potential answers to that problem. I don't know which one or which combination is best. But given the pace at which Generative AI is being adopted, I really hope we invest more time thinking about the gap in expertise we are creating. https://lnkd.in/gR9Pc9xj

This is an interesting perspective. We have all been hearing a lot lately about the growing capabilities of #ai and what roles are going to soon be made redundant because of that. There is a lot less discussion about what new areas will be in demand. What are your thoughts on this take? https://lnkd.in/gCmzpDcj At Connected Solution Experts, we focus on providing services in the 'Outer Loop' disciplines referenced in this article.

Here be dragons! A great perspective on the emerging use of AI in refactoring. https://lnkd.in/gs9SNA44

One of our core competencies at KRS | Khanyisa Real Systems is helping clients extend the useful life of older projects, and assist in transitioning to a modern solution. I've shared some thoughts on how we use of AI in this space.

Here are two articles about AI and how it relates to us as developers. Nice job Steven and Addy! https://lnkd.in/gdb9zwMP https://lnkd.in/gZMS4hyR

Writing unit tests for your code is important - but it can be time-consuming. So what if AI could help you out? Enter testing-specific LLMs. In this guide, Oluwadamisi shows how you can automate your testing with TestGen-LLM and Cover-Agent.