Hands-on CTO: why I still write code, and when it makes sense

A hands-on CTO keeps writing production code while running the company’s technology. That is what I have done at Blanco Brand since February 2024: on SuiteMedica, for example, my part is the architecture, leading the team and development. It works when the team is small and the software sits close to the customer’s operation, and it stops working when the CTO becomes the bottleneck.

By , CTO · Blanco BrandPublished 3 min read

What it means in my case

I have been CTO of Blanco Brand since February 2024. The role includes Keeper, a group company, and the technical platform I run has 25 deployed projects for clinics, mining, residential communities and other sectors. They are deployments of different sizes, not 25 products.

My part changes from project to project, and the cases I publish state it as it is:

  • On SuiteMedica: architecture, leading the team and development. I still write code.
  • On Alex Stewart: technical direction, development and project management.
  • On CARS Control: the architecture and the integration with the ERP.
  • On Keeper: I kept a single Flutter codebase, added the analytics layer and prioritised faster resident onboarding.

I started as a developer, became a tech lead and am now a CTO. I still do all three, depending on what the project needs.

Why I still write code

Architecture decisions come out better when the person making them knows the code that has to carry them. Routing SuiteMedica’s agent to language models through OpenRouter, or putting a read cache between CARS Control and the ERP, look small on a diagram and matter a lot day to day. I make those calls with more confidence when I implement them or review them closely afterwards.

Team size matters too. At Blanco Brand, delivery is organised around a stable team of three. In a team that size, a CTO who only manages leaves a large share of capacity unused, and writing code is part of how deadlines get met.

And then there is closeness to the customer. A good part of my work is sitting with the people who use a system before writing any code, which I describe in the note on Forward Deployed Engineers. That conversation pays off more when the person listening can change the code the same week.

What a hands-on CTO has to stop doing

The risk is well known: the CTO keeps the most interesting tasks, reviews everything and becomes the only person who understands parts of the system. At that point, writing code takes away from the team instead of adding to it.

What works for me, as a general rule:

  • Taking work that blocks nobody if I fall behind because of a meeting or an incident.
  • Letting the team’s rules live in continuous integration rather than in my review. On SuiteMedica every change goes through build, type-check, lint, unit tests and end-to-end authorization tests before it is accepted, whoever wrote it.
  • Writing down the decisions only I know about, so the next person does not have to ask me.

The numbers that belong to the CTO job

Writing code does not replace the responsibilities of the role. The figures I can back up from these years are about organisation and platform, not lines of code.

At Blanco Brand I reorganised delivery around a stable team of three. Development spend fell 63% and infrastructure spend 55%. Deployments went from one hour to five minutes, at 99.95% uptime. At Keeper, onboarding a resident now takes 40% less time, and the platform has 2,500 active users across more than 500 houses.

Those numbers come from CTO decisions: who to work with, which platform to run on, what to prioritise. Knowing how to code helped me make them with first-hand information.

When a CTO like this makes sense

It makes sense in small teams, where every person counts, and in companies whose software lives inside specific operations: a clinic, a lab, a residential community. There the distance between a technical decision and its effect on the operation is short, and it pays for the person deciding to walk all of it.

It makes less sense in organisations with many teams, where the CTO’s main job is coordinating, hiring and holding the strategy. In that setting, writing code tends to be an expensive hobby.

My full career history is on the about page.

If you lead a small team and are deciding how much code to write yourself, I would like to hear how you handle it. Write to me at hola@bcristi.com.

Let’s talk. If you want to discuss AI, healthcare, fintech or research, propose a collaboration or invite me to give a talk, write to me.

  • Asturias, Spain
  • CET hours
  • Spanish native, English C1

Also on WhatsApp · GitHub