Engineering maturity, cloud vs on-prem, and CTOs 💡
Monday Ideas — Edition #160
1) 🖥️ AI agents need a dedicated web browser
This idea is brought to you by today’s sponsor, Browserbase!
We need AI agents to browse the web, but we can’t just spin up Chrome and let them do their thing. We need browsing to be fast, scalable, and cheap. We need logs and recordings of past sessions, or debugging the browsing live.
Browserbase provides scalable and reliable browsing infrastructure for computer agents. You can learn more below 👇
2) 🔺 The Pyramid of Engineering Maturity
At the beginning of the year we released a wide industry report about what makes engineering teams successful.
Working through quantitative and qualitative answers, we traced a pyramid of maturity, which is made of three levels: Transparency → Intent → Speed
🖼️ Transparency — is about being aware of what’s going on, and creating the correct feedback loop to discuss and design improvements. Good communication, good retrospectives and continuous improvement are the #1 benefits that teams report from embedding data into their processes — and are foundational to everything the team does.
🎯 Intent — Transparency enables intent. The second clear trend that comes out is that teams that are intentional about parts of their engineering process are rewarded. Some of the clear wins are:
Allocating engineering time across a balanced portfolio of initiatives.
Creating better, participated requirements for features and projects.
Setting improvement goals for parts of the engineering process.
🔥 Speed — is the tip of the pyramid. Going fast is only helpful and sustainable when you are going in the right direction (intent) and you have a good feedback loop to steer your practices if needed (transparency). So, good teams are fast, but fast teams are not necessarily good. Shipping every day doesn’t magically turn you into an elite engineering team — but if you are an elite engineering team chances are you are shipping every day.
You can find the full report below 👇
2) ☁️ Cloud vs on-prem is complicated
When we hear people like DHH talking about "cloud vs on-premise", we often think in simple terms: it’s either full AWS, or servers blinking in our basement.
Reality, of course, is more complicated. Infra is made up of two fundamental layers:
🔩 Physical — the tangible stuff: the actual servers (compute, storage), networking gear, power, cooling, and physical security.
🪄 Virtual — the software and processes sitting on top of the physical hardware, to provision, deploy, and operate your applications. Think virtualization, container orchestration, deployment pipelines, monitoring, etc.
A full setup involves making choices at both of these layers. So let's look at the common options for each:
1) Physical infrastructure 🔩
Where do the actual servers live, and who owns them? You have four main options here:
Full on-premise — you own/lease the facility and hardware, managing everything yourself.
Co-location ("co-lo") — you own the hardware but rent space in someone else's datacenter.
Dedicated Servers — you rent entire physical servers, getting root access without managing hardware.
Cloud (IaaS) — you rent virtualized resources on demand from providers like AWS, paying for what you use.
2) Virtual Infrastructure / Management Layer 🪄
Once you have sorted out the physical layer — how do you manage and deploy applications on top of it?
Here are the main ways:
DIY / traditional stack — manual OS configuration, virtualization setup, and custom deployment scripts. Maximum control but high expertise needed.
Modern self-hosted tooling — Tools like Kamal, Dokku, and Caprover simplify application deployment with PaaS-like features, but still require underlying infrastructure management.
Cloud Managed Services (PaaS/FaaS/Managed K8s) — higher-level services from cloud providers that handle infrastructure management, letting teams focus on code. Examples include AWS Elastic Beanstalk, Lambda, and managed Kubernetes.
Hybrid Cloud Platforms — solutions like AWS Outposts and Google Anthos run cloud stacks on your hardware. Useful for specific needs like data residency, but complex and costly.
So, the choice isn't just cloud vs on-prem. It's first of all whether you want your hardware in your building, your hardware in someone else's building, rented hardware, or rented virtual resources.
And on top of that it’s how you manage and deploy to that hardware — DIY, using modern tooling like Kamal, using managed cloud services, or even a hybrid approach.
We published a thorough analysis of the whole infrastructure landscape in this recent piece 👇
3) 👑 A CTO’s primary team is the executive team
Last month I interviewed Joel Chippindale, professional CTO Coach and coach-in-residence in our community, about the challenges that CTOs face today.
One bit that stuck with me was when Joel talked about the need for CTOs to understand their role within the executive team.
Many early-career CTOs struggle with a fundamental mindset shift: seeing themselves primarily as members of the executive team rather than as representatives of the engineering team.
"It's not that […] my primary team is the engineering team and I'm there to support them and fight their battles in the leadership team. It's more that I'm a member of the leadership team. My engineering or technology team is my secondary team."
This shift transforms how CTOs approach their relationships with other executives:
🏢 Organization-first mindset — successful CTOs prioritize company success over engineering team advocacy.
🤝 Build executive relationships — they invest time in understanding each executive's challenges and priorities.
🌱 Cultivate trust — they demonstrate reliability, competence, and sincere care about the company's overall success.
When this shift doesn't happen, the effects ripple through the organization: teams begin mirroring executive conflicts, silos form, and the "us vs. them" mentality prevents effective collaboration across departments.
You can find the full interview below 👇
And that’s it for today! If you are finding this newsletter valuable, consider doing any of these:
1) 🔒 Subscribe to the full version — if you aren’t already, consider becoming a paid subscriber. 1700+ engineers and managers have joined already! Learn more about the benefits of the paid plan here.
2) 📣 Advertise with us — we are always looking for great products that we can recommend to our readers. If you are interested in reaching an audience of tech executives, decision-makers, and engineers, you may want to advertise with us 👇
If you have any comments or feedback, just respond to this email!
I wish you a great week! ☀️
Luca