“How many years’ experience do you have in agile software development?”
In 2014, I moved into embedded software development. Our project teams had around 30 people. We worked iteratively and incrementally, took safety and security seriously, and consistently released and delivered software.
Later, I learnt that this apparently wasn’t agile enough. We lacked daily stand-ups, sprints and PI planning.
Thankfully, Agile arrived to rescue us from “waterfall”.
From 2019 onwards, the software development organisations I worked in grew. So did the debates about scaling agile frameworks and applying them “correctly”. During that time, I didn’t witness a single safety approval within such an organisation.
Were the frameworks to blame? Probably not.
Can SAFe, LeSS and others help? Certainly.
Are they the key? No.
Now AI has entered the picture, bringing greater speed, new possibilities and added complexity. Producing code faster doesn’t automatically mean delivering a safe, secure and reliable product sooner.
Neither AI nor agile frameworks absolve us of responsibility for our products. Nor do they remove the need to get the fundamentals right.
We still need to understand our products, manage their complexity, understand regulatory requirements and bring together the right expertise.
That calls for less dogma and fewer buzzwords — and more product understanding, experience and pragmatism.
As development accelerates, how do you ensure your teams’ understanding of what it takes to build a product keeps pace?
#Agile #EmbeddedSoftware #AI #FunctionalSafety #ProductDevelopment #OrganisationalDesign