LLMs and Agile: Why Pit Them Against Each Other?
Just because we develop faster, does that mean we no longer need Agile? This idea betrays a misunderstanding of what Agile is.
Just because we develop faster, does that mean we no longer need Agile? This idea betrays a misunderstanding of what Agile is.
High-quality software is well-designed software before it is well-built software. Let’s take a look at how to get there.
The problem is that the benchmark (Waterfall) leads to software that doesn’t meet users’ needs…
Don’t ask, “How long will this take?” Instead, ask yourself if you can break your work down into even smaller pieces.
Some MVPs cost 10,000 euros; others, 10 million euros. The goal is the same, to acquire real users, but it all depends on the market.
User Story or technical task? Behind a battle of words lies the trade-off between today’s value and tomorrow’s.
“It’s good to learn from your mistakes. It’s better to learn from the mistakes of others.” Warren Buffett talks about investing; I talk about sprints.
Sequential methodologies do not work for business software, primarily because of tunnel vision. Is there an alternative? Agile.
If you had to keep just one thing from Agile, keep the retrospective. As a forum for continuous improvement, it paves the way for problem-solving.
The Walking Skeleton aims to lay a solid foundation for reaching the MVP while minimizing technical debt. It’s a step worth formalizing.
It would be a mistake to think functional design is the PM’s job alone: we think better together, working from concrete examples.
A PoC validates a marketing or technical idea, while an MVP aims to acquire your first users.
A project exists because we envision a goal to be achieved. A product has no end and evolves based on user feedback.
My main criticism of Scrum is that it says nothing about functional design. If you want it to work, it’s up to you to fill in the gaps.
“As a user, I want to connect to the database, so that I’m connected to the database.” Great… not. Surely we can do better than that.
Every feature added to a product increases its complexity and raises the risk of errors. So be absolutely certain of the value it adds.
Rather than making estimates, which are as accurate as a broken clock (which is right twice a day), break your work down into smaller tasks.
A simple account from someone who had the audacity to try to book a train ticket via the SNCF Connect website.
As Agile has spread, it has been distorted, becoming synonymous with micromanagement and absurd practices. Yet we need Agile now more than ever.