I keep hearing that AI has made Agile irrelevant. Okay. I'll throw in the towel. Let's dump Agile altogether.
So, instead, here is by far the best AI-focused way of working that I know:
First,
- The way we treat people and work with them is more important than processes and tools
- Working software is more important than any sort of documentation or other formalisms
- The best work happens when we work collaboratively with each other and our customers. The best environments are based on trust and communication.
- Planning only goes so far, and we need to adapt to what we learn as we learn it
Furthermore, I'd suggest that:
- Our highest priority is to satisfy the customer through early and continuous delivery of valuable software.
- The best way to build the right thing is to work small and deliver very frequently (at least daily), so that we can get feedback early and adjust based on that feedback.
- That we have to do that by working closely with business and product people. The code is secondary to defining and building the right thing.
- That we have to hire motivated people and get them the best tools that we can for them to do their job. They're the best people to tell us what tools they need.
- Communicating in real time eliminates lots of problems—delays, blockers, rework.
- We'll measure progress (and productivity) by how fast we can get valuable software into our customers' hands. Things like tokens/day, lines of code, and other measures of volume aren't useful. You are not more productive simply because you (or your AI helpers) type faster.
- We'll do our work in a humane environment, relaxed enough that we can come to work every day rested and able to do our best work. People matter.
- We'll focus on excellence and quality. Extensive testing, particularly when an LLM is involved, is essential. Great architecture underlies all great systems.
- We'll strive for simplicity, both in the code and in the product. We'll build exactly what our customers need and no more, and will determine their needs by talking to them.
- We'll trust the teams to figure out the best ways to go about their work.
- We'll constantly look at both what we're building and how we're building it, and continuously improve in both categories.
This is by far the best AI-focused approach that I know. Forget about all that "Agile" BS! 🙄
Another common question I’m answering working with scaling tech companies is…
Q. How much of your r&d spend should be focused on platform work?
A. 50%, and most teams are way off what it should be. Let me explain…🧵 (1/21)
What are the best universities for software engineering?
Congratulations to @UVA@CS_UVA for ranking #1 in CodeSignal's 2022 University Ranking Report. 🎉 See our top 50 US colleges and universities for software engineering: https://t.co/YUvH3x4hID
So you’re an engineering manager and thinking about becoming an individual contributor. If this is a viable option for you, I think it’s a splendid idea. Details in the thread.
On software quality: many teams obsess far too much on how to ship with close to no bugs to production:
but not enough about how to quickly spot these, and resolve them rapidly - doing this either automatically, or with a simple step.
Resilient products/platforms have both.
Exciting times at Cvent! We’re thrilled to announce our plans to go public and continue on our mission to transform meetings and events. https://t.co/h3o6UcdaJk
I had a fun time doing the #DevDcPodcast this week with @SparkTechTweets
https://t.co/9suJOUJ6jK
Watch the whole interview on https://t.co/Ctn0HbFLui or https://t.co/DZIi8AVQ80
Our team has a rule I like which is that every time somebody gets paged we create an incident. The first thing you do is triage and work out the impact. Many are negligible / minor, but then you have to discuss if you should have been paged.
A culture of refactoring is vital for successfully maintaining large codebases.
The more code changes, the easier it is to change.
Once code stops changing, it becomes virtually impossible to change. That inevitably leads to the dreaded rewrite and “second-system effect.”
🔥🔥🔥 @YoussefTaghlabi | ZDS 079 🔥🔥🔥
Super grateful to be able to share some of Youssef's story with y'all:
+ Innovation in US and Across Country Lines
+ Enterprise/Startup concerns
+ The Human Side of Engineers
+ Work in the Covid-Era
+ More!!!
https://t.co/p0AmiscLDj
🚗💨Accelerate your Bazel migration!
In this talk, @Jakeherringbone shares some learnings from rules_nodejs about how to adapt Bazel & rules to work with existing tools, keeping more parity with your legacy build.
Watch now 👉 https://t.co/tD5MeB99aF
#GoogleOSLive#Bazelday