Ole M.R. Rabanal

Most software fails before anyone writes a line of code

Wed, 28 Oct 2026 00:00:00 +0000

By the time a project is visibly in trouble, the decision that doomed it is usually months old. The code is rarely the problem. The problem is that the system was asked to do something nobody had stated precisely, for a business process nobody had mapped, under constraints nobody had written down.

This is why I keep ending up in the same kind of room: aging systems nobody fully understands, rewrites that stalled halfway, organisations caught between what the business wants and what engineering can deliver. These look like engineering failures from the inside. They almost never are.

Getting the thinking right up front is what makes AI build the right thing

Wed, 28 Oct 2026 00:00:00 +0000

The arrival of capable code generation did not change what makes software succeed. It changed the cost of the failure mode. Vague intent has always produced software that merely looks right; what is new is how quickly, and how plausibly, it can now be produced.

A model given a loose description will return something confident and complete. It will not stop to ask which of the four possible readings you meant, because nothing in the request signals that there were four. The output compiles, it demos, and it encodes an interpretation nobody chose.

Design so every developer delivers a little better

Wed, 28 Oct 2026 00:00:00 +0000

Most advice about building software assumes a team you do not have. Hire exceptional people; give them autonomy; get out of the way. It is good advice and it is rarely actionable. Real teams are assembled from who was available, at the rate the budget allowed, in the town the office is in.

The structure those people work inside, though, is something you can change. And it determines a surprising amount of what they are able to produce.

Modernising a data platform for a telecom provider

Fri, 01 Aug 2025 00:00:00 +0000

A legacy platform had grown across thirteen repositories, roughly sixty endpoints and around two hundred stored procedures. Nobody held the whole picture. The work was to recover the business logic buried in it, and to replace it with something that produced the same outcomes from far fewer moving parts.

Context

A leading telecom provider, from August 2025, as architect through Sonat AS.

It is a classic pattern: an old system is given more and more responsibility over time, and grows to meet it. Business rules ended up mostly in the database, as stored procedures with no version history, no tests, and no boundary between one domain and the next. The behaviour was real and the business depended on it, but it was not written down anywhere except in the implementation.

Recovering the business logic from an integration platform

Sat, 01 Oct 2022 00:00:00 +0000

A food company was building a new integration platform for products delivered by subcontractors. The system it was replacing was live, unversioned, and the only remaining record of a large amount of business logic.

Context

August to October 2022, through Sopra Steria.

The legacy system had to move, but a great deal of its business logic was not easy to reach. The deployed Java was unversioned — no history to read and no intermediate states to compare, so the reasoning behind any given behaviour was not recoverable from the code alone. Alongside it sat databases whose stored procedures also needed investigating and documenting. The logic had to be improved and optimised on the way out, not simply transcribed.

Trygg Tunnel — tunnel management for Statens Vegvesen

Fri, 01 Apr 2022 00:00:00 +0000

Statens Vegvesen and Vegdirektoratet both work on tunnels and had no shared system or process for managing them. Trygg Tunnel was built to give them one — and produced something along the way that had never existed before: a single view of the status of every tunnel in the country.

Context

March 2020 to April 2022, through Sopra Steria.

Statens Vegvesen plans, builds and maintains Norway’s national and county roads. Vegdirektoratet and Statens Vegvesen work together often, and needed a common system and process for managing tunnels. The web application covers the full tunnel lifecycle — planning, carrying out safety inspections, following up faults, and maintenance.

Production lines for the ParQSense sensor and base station

Sun, 01 Dec 2019 00:00:00 +0000

Software that builds physical products has an unusual property: the output is not a result on a screen, it is a device that leaves the building and cannot be patched afterwards.

Context

August 2018 to December 2019, at Q-Free ASA in Trondheim, in a supply chain team that owned both the hardware and the software for the production lines.

Two closely related projects, each producing one production line: one for single IoT parking sensor units, and one for the base station that multiple parking units communicate with.

Production line for the RSE651 tolling transceiver

Sat, 01 Sep 2018 00:00:00 +0000

The RSE651 is a DSRC transceiver — the unit in a tolling lane that communicates with the electronic toll tags in passing vehicles. This was the production line that manufactures it.

Context

August to September 2018, at Q-Free ASA in Trondheim.

The line is semi-automated: operators assemble and move products from one step to the next through the production process. The team was a project manager, a hardware electrical engineer, and a software engineer.

Curriculum Vitae

Mon, 01 Jan 0001 00:00:00 +0000