Skip to main content

We’re Still Engineers: Building for Outcomes

There is a lot of discussion about how AI is changing software development. We can generate code, tests and documentation faster than before. Some work that used to take hours can now start in minutes. This is a real change, and it affects how we work and how we think about our careers.

But when I ask myself what people will rely on me for, my answer is not only writing code. They will rely on me to understand the problem, make technical decisions, build and verify a solution, and learn from what happens after it reaches production. The tools change. Our responsibility for the outcome stays with us.

This matters because faster implementation does not automatically mean that the whole job is faster or easier. We may still wait for answers, access, data or reviews. And when a tool creates code quickly, we still need enough technical understanding to check whether it is correct, secure and useful in the real system. AI can help with the work, but it does not know the full business context and it does not take responsibility for the result.

For me, engineering starts before implementation. We need to ask who needs the outcome and what decision they are trying to make. How do they do it today? What evidence would change their mind? What information can we actually access? These questions can change what we build. A request for a complex platform may turn out to be a need for a clear estimate or a better way to make an early decision.

I saw this in a project I worked on with my team: an internal cloud infrastructure cost calculator, released in December 2024. People often had an idea for a solution but did not yet have all the details. They still needed an early budget estimate. We built a tool that could give a quick, consistent estimate, while making its assumptions and limits clear. Later, we explored whether a GPT could collect the needed information and fill in the calculator. That was an experiment, not the whole solution. The important part was understanding the decision people needed to make and building something useful for that stage.

The same thinking applies to modernising older applications. AI can help us explore code, identify dependencies and prepare changes. But an agent sees the code; it does not automatically understand every business rule, user expectation or operational risk around it. We need to assess the current behaviour, plan the work, make changes in steps, and validate that the application still does what people depend on. Good engineering means keeping control of the process and learning from evidence, not simply asking a tool to rewrite everything.

This is why I believe we need more people who are willing to build and try things. A small prototype can make an unclear question visible. It gives people something real to react to. We can learn, improve it, or decide to stop. Not every idea needs a large project, but ideas need a way to meet reality.

AI will change our daily work, and our roles will keep changing too. We should learn the tools and use them thoughtfully. At the same time, we should keep practising curiosity, judgement and care for the people who use what we build. At the end of the day, we are still engineers. We solve problems, we build, we verify, and we own the outcomes.

Comments

Popular posts from this blog

How to audit an Azure Cosmos DB

In this post, we will talk about how we can audit an Azure Cosmos DB database. Before jumping into the problem let us define the business requirement: As an Administrator I want to be able to audit all changes that were done to specific collection inside my Azure Cosmos DB. The requirement is simple, but can be a little tricky to implement fully. First of all when you are using Azure Cosmos DB or any other storage solution there are 99% odds that you’ll have more than one system that writes data to it. This means that you have or not have control on the systems that are doing any create/update/delete operations. Solution 1: Diagnostic Logs Cosmos DB allows us activate diagnostics logs and stream the output a storage account for achieving to other systems like Event Hub or Log Analytics. This would allow us to have information related to who, when, what, response code and how the access operation to our Cosmos DB was done. Beside this there is a field that specifies what was th...

Can User Technical Profile Influence AI Architecture Decisions?

In the last few days, I ran a small experiment on Devin/Windsurf/Cascade, BMAD, and LLM-based architecture decisions. The initial concern was simple: if an AI coding assistant knows something about the user, for example, that the user is a Java developer or a .NET developer, can this information influence the architecture and technology stack that the AI will propose? This question is important because many companies are starting to use AI tools not only for coding but also for product discovery, architecture, documentation, and technical decision-making. If the AI is influenced by the developer's personal profile, the result may not be fully neutral. It may look like a business or architecture decision, but in reality, it may be partially shaped by the context of the person using the tool. The test was done using BMAD with Devin/Windsurf/Cascade. The scenario was a unified field service operations platform. The task was to let BMAD refine the product, define the architecture and s...

ADO.NET provider with invariant name 'System.Data.SqlClient' could not be loaded

Today blog post will be started with the following error when running DB tests on the CI machine: threw exception: System.InvalidOperationException: The Entity Framework provider type 'System.Data.Entity.SqlServer.SqlProviderServices, EntityFramework.SqlServer' registered in the application config file for the ADO.NET provider with invariant name 'System.Data.SqlClient' could not be loaded. Make sure that the assembly-qualified name is used and that the assembly is available to the running application. See http://go.microsoft.com/fwlink/?LinkId=260882 for more information. at System.Data.Entity.Infrastructure.DependencyResolution.ProviderServicesFactory.GetInstance(String providerTypeName, String providerInvariantName) This error happened only on the Continuous Integration machine. On the devs machines, everything has fine. The classic problem – on my machine it’s working. The CI has the following configuration: TeamCity .NET 4.51 EF 6.0.2 VS2013 It see...