During September , we hosted the “Robot Framework & AI” meetup at our Finland Sogeti office. The evening was filled with insightful talks about the future of Generative AI (Gen AI), Robot Framework, and test automation, sparking many fruitful discussions on these topics. In this blog post, I want to provide a brief recap of the key highlights from the event.
As the clock hit 4:30 PM, attendees began to gather in our lounge. By 5:00 PM, with a full house, it was time to kick off the event. I had the pleasure of welcoming everyone, introducing myself, and acknowledging our five hosts who ensured the evening ran smoothly..
Our first speaker, Miikka Solmela, Head of the Robot Framework Foundation, introduced the Foundation and shared updates from the Robot Framework community. AI was a central theme, with Miikka discussing the ongoing development of the Robot Framework AI library andthe Robot Framework Chat Bot, which contains knowledge about available libraries and keywords.
Next, it was my turn to present Sogeti’s Gen AI Amplifier for Software & Quality Engineering. I began by emphasizing the transformative moment we are currently experiencing with Gen AI and how it can significantly boost productivity. Many AI features are now being integrated into real-world applications. I then introduced the Gen AI Amplifier and explored its use cases, which covers the software development lifecycle. One key example was how it can generate various documents essential to software development & testing
The core of my presentation was a live demo of Robot Framework script generation. I started with a test case document in Word format that outlined test case requirements and steps. After feeding this document into the Gen AI Amplifier, the system generated a well-structured Robot Framework test script. I also demonstrated a previously generated script, explaining how minor modifications allowed it to fully automate the test case from start to finish.
I concluded my talk by discussing the broader implications of Gen AI for test automation developers. I suggested that our roles may shift from primarily being code producers to more of a code verification role, acting as analysts who decide which AI-generated content to keep and which to regenerate. While we are still in the early stages of the Gen AI transformation, I believe many exciting developments lie ahead.
The next speaker was Joonas Peuralinna from Relex, who presented on using Gen AI to enhance Robot Framework documentation. Joonas showcased a VSCode extension he developed that makes API calls to OpenAI to generate documentation from keywords. What stood out was how Joonas built the extension with the assistance of ChatGPT, despite having no prior experience with TypeScript. His presentation highlighted that documentation generation is an excellent use case for Gen AI—since small semantic errors won’t lead to unexecutable code, only minor grammatical issues that can be easily corrected.
Following the presentations, we took a short break to enjoy some snacks and refreshments. We then divided the attendees into five small discussion groups, each centered around a specific question. After 30 minutes of engaging discussions, each group presented their key takeaways to everyone. The range of topics was broad, but common themes included concerns about how junior developers will learn in this new AI-driven landscape, the future of Large Language Model development, and strategies for integrating these models into software pipelines.
From my perspective, the event was a great success. In my opening remarks, I said that I would consider the event successful if each attendee left with at least one new idea about Gen AI and Robot Framework. Personally, I left with far more than just one idea!
Thank you again to everyone who attended, and I look forward to seeing you at the next meetup!