<?xml version="1.0" encoding="utf-8" standalone="yes"?><rss version="2.0" xmlns:atom="http://www.w3.org/2005/Atom"><channel><title>Blog on CTOMultiplier</title><link>https://ctomultiplier.com/tags/blog/</link><description>Recent content in Blog on CTOMultiplier</description><generator>Hugo</generator><language>en</language><lastBuildDate>Mon, 23 Mar 2026 21:05:53 +0100</lastBuildDate><atom:link href="https://ctomultiplier.com/tags/blog/feed.xml" rel="self" type="application/rss+xml"/><item><title>How to Adopt AI in a Software Team</title><link>https://ctomultiplier.com/how-to-adopt-ai-in-a-software-team/</link><pubDate>Sun, 16 Nov 2025 13:15:38 +0000</pubDate><guid>https://ctomultiplier.com/how-to-adopt-ai-in-a-software-team/</guid><description>&lt;p&gt;Does your AI adoption strategy consist of giving out Copilot, Cursor, or Claude licenses and letting them figure out how to use them?&lt;/p&gt;
&lt;p&gt;Over the past year, I&amp;rsquo;ve found this to be a widespread strategy in software companies, whose teams face high workloads and, due to lack of knowledge and time, see AI as just another technology that can be learned self-taught by each developer.&lt;/p&gt;
&lt;p&gt;When I start working with a client who has followed this approach, the first thing I do is measure the actual use and impact the initiative has had, and what I observe is the following:&lt;/p&gt;</description></item><item><title>Why We Moved Away from WordPress</title><link>https://ctomultiplier.com/why-we-moved-away-from-wordpress/</link><pubDate>Fri, 03 Oct 2025 11:35:14 +0000</pubDate><guid>https://ctomultiplier.com/why-we-moved-away-from-wordpress/</guid><description>&lt;p&gt;Two years ago, when this website was created, WordPress was practically the default option for creating landing pages and blogs. Being a visual tool (WYSIWYG - What You See Is What You Get) with a large community, it has many templates and plugins that allow you to build something quickly and maintain it without worrying too much about the details. For all these reasons, we chose WordPress to create the first version of this website.&lt;/p&gt;</description></item><item><title>Guide to robust development with AI: Cursor and Github Copilot</title><link>https://ctomultiplier.com/guide-to-robust-development-with-ai-cursor-and-github-copilot/</link><pubDate>Wed, 14 May 2025 15:48:12 +0000</pubDate><guid>https://ctomultiplier.com/guide-to-robust-development-with-ai-cursor-and-github-copilot/</guid><description>&lt;p&gt;AI code assistants have recently spread due to the power they offer to speed up software development. However, learning to use them is not as simple as writing a simple prompt. The learning curve to fully take advantage of what they offer is not as small as we might think. If we underestimate it, we risk a) producing poor quality code with errors and security flaws or b) making limited use of the capabilities and obtaining fewer benefits than those available.&lt;/p&gt;</description></item><item><title>Is Coaching Useful for a CTO?</title><link>https://ctomultiplier.com/is-coaching-useful-for-a-cto/</link><pubDate>Sun, 28 Jul 2024 21:20:05 +0000</pubDate><guid>https://ctomultiplier.com/is-coaching-useful-for-a-cto/</guid><description>&lt;p&gt;Last year, I enrolled in a coaching course. I decided to do this after realizing that my work as a Fractional CTO involved a lot of mentoring, helping engineers transform into managers and CTOs, and helping teams improve their way of working. Although I feel that my mentoring style has generally been effective, I have noticed that it hasn&amp;rsquo;t always worked with some people and teams. At the same time, several people in my circle told me about coaching and how it had helped them professionally in working with teams and individuals. For all these reasons, I decided to enroll in a course with the expectation of improving my skills in these areas, but what I didn&amp;rsquo;t imagine was that coaching would bring me more than I expected and help me rethink my style as a CTO.&lt;/p&gt;</description></item><item><title>Groundhog day</title><link>https://ctomultiplier.com/groundhog-day/</link><pubDate>Thu, 25 Jan 2024 10:35:32 +0000</pubDate><guid>https://ctomultiplier.com/groundhog-day/</guid><description>&lt;p&gt;It will soon be 4 years since the world changed and remote working became more popular. Initially adopted as a way of limiting social contact, it was a very effective instrument. We distanced ourselves socially but at the same time we were able to continue working and being productive (sometimes even more productive), without the world coming to a standstill. Suddenly we went from spending up to 2 hours commuting to the office to being able to wake up ten minutes before the start of the working day and start work in our pyjamas, or at least in a pyjama shirt and trousers. And this in turn opened up a whole range of possibilities: we could work for companies in other geographies, have meetings with clients without having to travel, receive packages from Amazon and the washing machine technician and so on. What&amp;rsquo;s more, we suddenly had more time that we could spend on going to the gym, leisure, hobbies, enjoying time with friends and family, or even working more.&lt;/p&gt;</description></item><item><title>¿Agile vs Waterfall?</title><link>https://ctomultiplier.com/agile-vs-waterfall/</link><pubDate>Wed, 24 Jan 2024 08:42:11 +0000</pubDate><guid>https://ctomultiplier.com/agile-vs-waterfall/</guid><description>&lt;p&gt;In a world where agility has become the norm for managing software projects, wondering if Waterfall and traditional project management is an option seems unimaginable, however the other day talking to a manager, he asked me if for implementation projects and customisation of a product in customers, traditional project management can be more interesting than using an agile methodology. And the truth is that it seemed like a good opportunity to write about what Agile brings to the table compared to other more traditional models, aimed at a management audience that is not so familiar with agility.&lt;/p&gt;</description></item><item><title>Systems Thinking</title><link>https://ctomultiplier.com/systems-thinking/</link><pubDate>Mon, 13 Nov 2023 18:11:56 +0000</pubDate><guid>https://ctomultiplier.com/systems-thinking/</guid><description>&lt;p&gt;Occasionally I work with clients who have a major problem in delivering on time and to the agreed quality. When I start to investigate the causes I find that when I talk to the product team they say the problem is that engineering is not fast enough and/or they produce too many defects, and when I talk to engineering they tell me that they get incomplete specifications from product, which means they have to spend more time developing, and sometimes they have to undo changes because they were not what the customer wanted. It is common for these organisations to see themselves as a set of silos, and to believe that the solution lies in &amp;ldquo;fixing&amp;rdquo; one of these silos (e.g. engineering). But this is a reductionist view of the problem, and rarely helps to solve it. To address the root of the problem, we need to rethink how we understand the organisation, and this is where systems thinking can help. As Einstein said: &amp;ldquo;We cannot solve problems by thinking in the same way as when we create them&amp;rdquo;.&lt;/p&gt;</description></item><item><title>The impact of culture on performance</title><link>https://ctomultiplier.com/the-impact-of-culture-on-performance/</link><pubDate>Mon, 13 Nov 2023 17:44:00 +0000</pubDate><guid>https://ctomultiplier.com/the-impact-of-culture-on-performance/</guid><description>&lt;p&gt;This week I want to talk about a podcast from techlead journal (link at the end), in which they interview the author of the book Wrong fit, right fit. In the podcast, the author (André Martin) talks about company culture as a determining factor for the performance and happiness of an employee. Quote &lt;strong&gt;a Gallup report that says $8.8 trillion is lost due to employees&amp;rsquo; lack of engagement with their companies&lt;/strong&gt;. André will equate working in a company with which you don&amp;rsquo;t have a cultural fit to writing with your left hand (or right if you are left-handed). You can do it, and you will produce&amp;hellip;but the quality will be much worse and you will feel unfulfilled.&lt;/p&gt;</description></item><item><title>How to conduct effective meetings</title><link>https://ctomultiplier.com/how-to-conduct-effective-meetings/</link><pubDate>Thu, 26 Oct 2023 10:53:18 +0000</pubDate><guid>https://ctomultiplier.com/how-to-conduct-effective-meetings/</guid><description>&lt;p&gt;There is often talk about whether meetings are worthwhile, and whether they should be abolished altogether because they are a waste of time. Personally I find this view somewhat extreme, but I understand that it responds to the fact that in many cases the way we conduct meetings is effectively a waste of time and money, and a source of frustration and demotivation. Throughout my career I have worked with quite a few companies, and in all of them I have participated in unproductive meetings, and this seems to be a constant no matter who you talk to. It&amp;rsquo;s hard to find someone who, when you ask them about how they do meetings in their company, doesn&amp;rsquo;t complain.&lt;/p&gt;</description></item><item><title>Why do ChatBots hallucinate?</title><link>https://ctomultiplier.com/why-do-chatbots-hallucinate/</link><pubDate>Thu, 05 Oct 2023 10:28:23 +0000</pubDate><guid>https://ctomultiplier.com/why-do-chatbots-hallucinate/</guid><description>&lt;p&gt;Those of you who have used ChatGPT, Google Bard or similar, have probably found that sometimes these chatbots make up the answers to our questions. This is what is commonly known as hallucinations.&lt;/p&gt;
&lt;p&gt;To understand why they happen, the first thing is to understand at a very basic level how these chatbots work. The fundamental building block is the language models (LLMs) &lt;em&gt;large language models&lt;/em&gt;). These models are trained on large amounts of data, such as web pages on the internet and books in the public domain, among others. The task of LLMs is to try to predict the next word or sequence of words from a text that the user enters. For example, if we ask a question, the model predicts the words right after that question. As the model has been trained on millions of documents, it is likely that in one (or many) of those documents it has seen a similar question, along with the answer. Roughly speaking, the LLM works like a statistical model, first during its training it learns the probability that two or more words go together, and then during its use it uses this probability to predict the next sequence of words;&lt;/p&gt;</description></item><item><title>Do you know what DPE is?</title><link>https://ctomultiplier.com/do-you-know-what-dpe-is/</link><pubDate>Wed, 27 Sep 2023 07:23:39 +0000</pubDate><guid>https://ctomultiplier.com/do-you-know-what-dpe-is/</guid><description>&lt;p&gt;DPE stands for Developer Productivity Engineering and is the name of a new discipline that aims to improve developer productivity through automation, observability and tool enhancement.&lt;/p&gt;
&lt;p&gt;The work of a developer when programming has three phases: code -&amp;gt; build -&amp;gt; test, the developer repeats this sequence tens or even hundreds of times a day. And in many cases the build and test phases can take in the order of minutes. For example: if the build of an application takes 5 minutes and the developer does 10 builds per day, that&amp;rsquo;s 50 minutes that the developer has to wait per day. Similarly, the testing phase can increase the developer&amp;rsquo;s waiting time. Multiply that by the number of developers in a company and the cost is significant.&lt;/p&gt;</description></item><item><title>What does a Fractional CTO do?</title><link>https://ctomultiplier.com/what-does-a-fractional-cto-do/</link><pubDate>Wed, 13 Sep 2023 15:01:31 +0000</pubDate><guid>https://ctomultiplier.com/what-does-a-fractional-cto-do/</guid><description>&lt;p&gt;A Fractional CTO is a professional who has worked as a CTO (&lt;em&gt;Chief Technology Officer&lt;/em&gt;) in technology companies or startups and now offers this expertise as a freelance or consulting service. This model allows companies to flexibly bring in an experienced CTO to help with their technology and product strategy. There are many reasons why a company might consider hiring a Fractional CTO, such as filling the CTO role in an early-stage startup, mentoring and supporting a manager in their promotion to CTO, conducting &lt;em&gt;due diligence,&lt;/em&gt; or strengthening the management team on a specific project.&lt;/p&gt;</description></item><item><title>Startup Mistakes: Hiring a full-stack developer as CTO</title><link>https://ctomultiplier.com/startup-mistakes-hiring-a-full-stack-developer-as-cto/</link><pubDate>Wed, 06 Sep 2023 13:36:42 +0000</pubDate><guid>https://ctomultiplier.com/startup-mistakes-hiring-a-full-stack-developer-as-cto/</guid><description>&lt;p&gt;Some startups hire a full-stack developer for the CTO position, because they do not have the budget for an experienced CTO. There is a belief that in the early stages a developer can take over the CTO position, and learn as they go along. There are several reasons for this. On the one hand, many startups start with very little money, which makes it impossible to pay the salary of an experienced CTO. On the other hand, there is a lack of knowledge about the responsibilities and value of a CTO.&lt;/p&gt;</description></item><item><title>Startup Strategy 101: 10 Tech and Product Errors to Avoid</title><link>https://ctomultiplier.com/10-mistakes-startups-make-with-their-technology-and-product-strategy/</link><pubDate>Wed, 06 Sep 2023 13:15:38 +0000</pubDate><guid>https://ctomultiplier.com/10-mistakes-startups-make-with-their-technology-and-product-strategy/</guid><description>&lt;p&gt;When you start a startup you have a blank sheet of paper: the possibilities are endless and while it is true that there is no 100% effective manual on how to get started, there are recommendations about what usually works and what doesn&amp;rsquo;t. In this post I want to talk from my experience working in several startups, about common mistakes I have observed when defining the technology and product strategy. Of course, it is not a complete list, and you can surely find exceptions for most of the points, but I think these are decisions that in many cases can be considered as wrong or at least not optimal. In the startup world, mistakes are often referred to as learning, but stumbling twice on the same stone is no longer learning&amp;hellip;&lt;/p&gt;</description></item><item><title>Norway: The country of self-service?</title><link>https://ctomultiplier.com/norway-the-country-of-self-service/</link><pubDate>Wed, 23 Aug 2023 10:09:13 +0000</pubDate><guid>https://ctomultiplier.com/norway-the-country-of-self-service/</guid><description>&lt;p&gt;This summer I have been traveling around Norway and I have been surprised to see how most of the stores adopt to a greater or lesser extent the self-service model to offer their services. For example, in coffee shops it is common to find coffee machines where the customer makes his own coffee. At the airports there are check-in and bag drop machines, and specifically at Evenes airport, this is the first time I have seen an airport without people at the bag drop counter. We even boarded the plane with a machine&amp;hellip;&lt;/p&gt;</description></item><item><title>Relationship between team organization and application architecture</title><link>https://ctomultiplier.com/the-relationship-between-team-organization-and-architecture/</link><pubDate>Fri, 04 Aug 2023 12:06:03 +0000</pubDate><guid>https://ctomultiplier.com/the-relationship-between-team-organization-and-architecture/</guid><description>&lt;p&gt;This relationship is known as Conway&amp;rsquo;s law, which was made by Melvin Conway in 1967 and reads as follows: &amp;ldquo;Organizations that design systems [&amp;hellip;] are limited to producing designs that are copies of the communication structures of these organizations.&amp;rdquo;&lt;/p&gt;
&lt;p&gt;Communication within the team is frequent and fluid, with regular planning meetings, dailies, retrospectives and demos. This allows the team to have a high level of alignment and agile coordination, which is fundamental to the development of a component. If you suddenly decide to split this team into 2 teams, there is a disconnect in communication. Communication with other teams is much less frequent, as one of the ideas of teams is to isolate them from externalities so that they can focus on their scope. This reduction in communication between teams makes the task of maintaining a single component more complex. In these cases, there are two options: 1) Split the component into 2, one per team, this allows the teams to decouple and work more autonomously. 2) to keep a single component, which implies setting up a series of coordination mechanisms and meetings between the teams, with the overload that this entails. In addition, the teams are interdependent, which reduces speed and autonomy and affects productivity (since they have to work so closely together, we could almost say that for practical purposes there is still only one team).&lt;/p&gt;</description></item><item><title>Guide to do Code Reviews in your team</title><link>https://ctomultiplier.com/guide-to-do-code-reviews-in-your-team/</link><pubDate>Tue, 25 Jul 2023 15:42:32 +0000</pubDate><guid>https://ctomultiplier.com/guide-to-do-code-reviews-in-your-team/</guid><description>&lt;p&gt;Code reviews are a practice that has become widespread in recent years, where one or more developers review the new code implemented by another colleague, with the aim of detecting code quality issues, bugs, vulnerabilities, bad practices, etc&amp;hellip; . This allows feedback loops to be shortened, which we know is very beneficial because the later a problem is discovered, the higher the cost of fixing it and the greater the potential business impact.&lt;/p&gt;</description></item><item><title>Is your Scrum suffering from low productivity? Your problem could be WIP</title><link>https://ctomultiplier.com/is-your-scrum-suffering-from-low-productivity-your-problem-could-be-wip/</link><pubDate>Thu, 20 Jul 2023 09:18:13 +0000</pubDate><guid>https://ctomultiplier.com/is-your-scrum-suffering-from-low-productivity-your-problem-could-be-wip/</guid><description>&lt;p&gt;It has been proven that the more work in progress a person or a team has, the more time it takes to close tasks. On the one hand, if we are doing several tasks at the same time and we deliver them at the same time, the delivery time will be the sum of the time of doing the two tasks. However, if we work on the tasks in sequential order, the first task will be delivered earlier, and the average delivery time will also be lower.&lt;/p&gt;</description></item><item><title>AI is not the product</title><link>https://ctomultiplier.com/ai-is-not-the-product/</link><pubDate>Fri, 14 Jul 2023 17:20:07 +0000</pubDate><guid>https://ctomultiplier.com/ai-is-not-the-product/</guid><description>&lt;p&gt;Lately I have spoken to several founders who have told me that people have been asking them to do things with AI for their products/services. Clearly there is a hype, and a lot of attention is being focused around AI. In addition, there seems to be a wave of powerful investments in startups with AI in their name, which feeds the hype.&lt;/p&gt;
&lt;p&gt;If I can give my two cents on the subject, based on my experience working in a startup that used AI to build products, I will say:&lt;/p&gt;</description></item><item><title>When abstractions leak</title><link>https://ctomultiplier.com/when-abstractions-leak/</link><pubDate>Thu, 15 Jun 2023 18:47:25 +0000</pubDate><guid>https://ctomultiplier.com/when-abstractions-leak/</guid><description>&lt;p&gt;In the software world we constantly use abstractions to make our work easier, as they abstract us from other domains and significantly reduce the complexity we have to deal with. An example of abstraction is ORMs. An ORM allows us to abstract the application from the database being used. Instead of having to write queries (specific to the database engine) to access the data, the ORM allows us to use the programming language we are using to code the application, also to access the data. Anyone who has used ORMs knows that in many cases the abstraction works well and brings a lot of value, but when the application requires us to get multiple related entities, among other examples, is when we start to see the limitations. In these cases, it is common that ORMs do not always generate the optimal queries, or that they require us to write excessively complex code, so that sometimes we end up writing queries directly in the language of the underlying DB. This means setting aside abstraction for such cases.&lt;/p&gt;</description></item><item><title>The cost of being reactive vs. being proactive</title><link>https://ctomultiplier.com/the-cost-of-being-reactive-vs-being-proactive/</link><pubDate>Sun, 04 Jun 2023 16:30:16 +0000</pubDate><guid>https://ctomultiplier.com/the-cost-of-being-reactive-vs-being-proactive/</guid><description>&lt;p&gt;On countless occasions we hear that we need to be more proactive. In other words, taking the initiative and anticipating problems. But being proactive almost always comes at a cost, which is paid in advance, at best a little time investment, at worst a lot of time and money. We often look at the cost of being proactive, but overlook the cost of not being proactive. Especially when we see that the cost of being proactive is high, we often dismiss it outright, without seeing the other side of the coin.&lt;/p&gt;</description></item><item><title>A short reflection on AI and low cost</title><link>https://ctomultiplier.com/a-short-reflection-on-ai-and-low-cost/</link><pubDate>Wed, 03 May 2023 16:32:00 +0000</pubDate><guid>https://ctomultiplier.com/a-short-reflection-on-ai-and-low-cost/</guid><description>&lt;p&gt;It seems that one of the debates surrounding AI centres on whether this technology will eliminate a large number of jobs by matching or surpassing human capabilities. Although in almost no discipline today does AI match a human expert, there are already thousands of products that integrate AI, and the number is growing daily. This is a phenomenon that has exploded in the last year with the popularisation of generative AI models, and especially with the emergence of ChatGPT, the brainchild of OpenAI.&lt;/p&gt;</description></item><item><title>Revisiting Agile Methodologies: XP, Scrum, Kanban and Scrumban</title><link>https://ctomultiplier.com/revisiting-agile-methodologies-xp-scrum-kanban-and-scrumban/</link><pubDate>Thu, 16 Mar 2023 10:47:00 +0000</pubDate><guid>https://ctomultiplier.com/revisiting-agile-methodologies-xp-scrum-kanban-and-scrumban/</guid><description>&lt;p&gt;It&amp;rsquo;s been almost 10 years since I started working with &lt;em&gt;agile&lt;/em&gt;. I remember my first experience was when I was working in southern France for Amadeus. At that time, the company&amp;rsquo;s management proposed changing and updating the way development teams worked, and carrying out an adoption of agile methodologies, mainly using the Scrum framework. In my team, they offered me Scrum Master training, which lasted 3 days, after which I had to take an exam to get certified. And that&amp;rsquo;s how I became a certified Scrum Master.&lt;/p&gt;</description></item><item><title>How to Do Load Testing for APIs</title><link>https://ctomultiplier.com/how-to-do-load-testing-for-apis/</link><pubDate>Fri, 03 Mar 2023 11:09:00 +0000</pubDate><guid>https://ctomultiplier.com/how-to-do-load-testing-for-apis/</guid><description>&lt;p&gt;If you&amp;rsquo;ve ever developed a web application designed to handle a significant volume of users, you&amp;rsquo;ve faced the question of determining what resources your application needs to handle expected traffic with desired service levels. Performance and load testing allow you to understand how your application will behave in different usage scenarios by answering the following questions:&lt;/p&gt;
&lt;ul&gt;
&lt;li&gt;
&lt;p&gt;How many resources (CPU, RAM, disk, bandwidth, GPU, etc.) does my application consume to process traffic?&lt;/p&gt;</description></item></channel></rss>