Approach

Quest for Quality

If I had to reduce everything I am trying to achieve to one word, then this one word would be "quality".

Devotion to quality as the main objective makes things easier by eliminating a number of choices. Certain things must not be done if you want a high-grade result. Other things must be done at all costs.

Quality is not just an attribute of the product. Quality must be inherent to the processes that will lead you to the product.

Flawed methods might occasionally produce good results. Unfortunately, those results will not be repeatable.

I keep looking for methods that make the work simpler, the software more robust, and the process repeatable.

I do not favor particular tools, programming languages or platforms. They are means to an end. Flexibility is one of the central requirements of software construction.

I choose techniques that will help me achieve exactly what my clients want: The agreed functionality at the agreed time for the agreed price. If I can provide this, I am gleeful.

Errors are a good thing

I love to find errors! Every error I help you find and fix is one your customers won't have to discover. It is very rewarding. I like that concept. Even more gratifyingly, errors give us the opportunity to improve process.

You might ask: Where is the connection between a solitary error and process?

Errors do not just pop up from nowhere. They are created, created by human beings, myself, you or the guy who is on vacation right now (it is his fault most of the time). AI can introduce errors, too. It can also help us find them.

Seriously, none of the people who make mistakes and create errors are to blame. To err is human and should never be criticized. What it really tells us is that somewhere in the development process a gap exists. Part of the process either encourages the creation of the error or prevents its timely detection.

I am curious. The project team and I investigate and research until we pinpoint the cause. And then we do not merely fix the error but we also patch the procedural safety net that should have prevented it in the first place. We can now be confident that an entire class of errors will not come back to bug us. A good thing.

Whether code is written by people or generated by AI, we need a sound understanding of software construction and architecture. If an error reveals a gap in that understanding, we need to provide training, designate a senior engineer as a coach or strengthen design and code review.

If the error was created through ambiguous requirements, we need to find a method or tool that will improve consistency and clarity.

If we find out that the source of the error lies in system design, we have to find ways to verify and validate our design in earlier project stages.

I will not rest until I know what went wrong and why. Every error, every human mistake will eventually, permanently, improve the development process, will bring us one step closer to our goal:

Zero errors, superb quality.