A big step in evolution, not
I spend an hour or two on LinkedIn every day. LinkedIn also recommend when I read more posts about sharing my experiences using AI code generation tools to solve their needs. It is impressive to see these stories of success – one person produces a game in two hours, the other applys for ten.
I started my career as a software engineer 34 years ago. I have spent more time working with software engineers in my entire career than any other character. I often hear about the future of software engineers. Many people outside the software development assume that the software engineers will be obsolete. Meanwhile, most practicing engineers do not see it and are often denied. They argue that AI tools are not yet able to replace them – and the future will not be in the near future.
After spending decades in the software, I felt forced to share my thoughts and predictions on the subject.
Considering his initial career as an individual partner in the 90s, I remember working on various projects using Z80, Motorola 68K, I386 Assembly, C, and C ++. Despite being a programmer, I spent only 25 – 40 % of my time coding. A few days, I coded intensely for 80 – 110 % in my working hours, but they were rare. In many other days, I didn’t write a line of code. If I am an average of it, less than half of my time was spent in coding.
At this time, most projects followed the waterfall model with the classified teams. My team lead or manager will assign me a task or module, and I will develop it based on their requirements and expectations.
Interestingly, I felt guilty for coding only 3-4 hours a day, assuming that full-time job means coding for up to 8 hours. It made me anxious and restless as I change with every new job, I was coding less time than the previous job. He left me four jobs within five years.
By the end of 1995, I decided not to burden my employers and became a business. Once I started to manage customers’ needs and expectations, I had more work that I could handle. I noticed that the best way to manage the work was to include key members of the team directly with users. As a result, I started practicing Fatali methods in early 1996.
Some companies and teams provide engineers with a saudocood or wire frame and expect them to translate these people into the working code. In my view, these are software programmers, not software engineer.
The role of a software engineer begins with understanding, explanation and scoping requirements, then designing, development, testing, integration, issuance, issuance, and sometimes deploying solutions.
There are also easy applications or tools designed for highly specific purposes. In the past, we used drag and drop, nine codes, or low code tools to create such solutions. AI tools can replace some of these WYSIWYG builders, nine codes and low code tools.
However, most software engineers work on large -scale projects or products that require deployment, support, maintenance, updates and big new release over many years.
Please keep in mind this context as you read.
Since 1996, I have been managing product release and software engineering teams. When I think about what a successful software engineer makes, I realize that coding is not the most important skill.
The success of the software engineer depends on their ability:
Listen to the requirements and analyze them well
Ask fully relevant questions to understand the copuse scope, limits and exceptions
• Understand how their work fits in a large project or product life cycle
The approach and design of a solution depends on the engineer’s exhibition, awareness, knowledge, and skills to solve the problem.
Interestingly, big software engineers are expected to be “slow”. They do not immediately start coding after understanding the requirements. Instead, they identify problems and solutions – whether the industry’s best methods, company plans, or previous work. They focus on maximizing reuse, evaluating whether they can increase existing solutions, make a reflect or custom.
Recently. , Their responsibility is not just to provide solutions, but also to ensure long -term maintaining and supporting.
Related to Living Software, software engineers should update their knowledge about the latest technologies, tools, processes, excellent methods, third -party libraries and APIS.
Over the past 6-7 years, we have adopted BDD (behavioral growth), TDD (test-driven growth), and CI/CD (continuous integration and permanent deployment). These methods have made the role of software engineers even more important.
When I see how the engineers spend their time over a year, only 20 to 40 % of coding is only in coding. Is dedicated to the rest of the time:
Requirement Discussions
use Case case development and reviews
• API design and final form
• Functional design
Re -reacting, reflecting, and indicating the customization opportunities
AG Forty Stories and Epic
Test automation for DTD
Daily Daily Stand Ops
• Critical Big Fixes and Instant Production Support
If AI tolls work exactly, they can significantly reduce time engineers spending coding in the limits of 20 – 40 %.
In order to get the most out of AI tools, software engineers must say:
• Craft precise indicators
AI-AI-Infield Code Review
Test it against requirements requirements
OUTS indicates a better output to improve the output
In addition, engineers have to make sure that the AI-infield code is linked to a safe, maintained, and project guidelines.
AI tools enable software engineers to automatically produce high quality products to repeated coding tasks. Just as social media has turned millions of readers into authors, so can many AI code readers become a great engineer-even though they have strong understanding, critical thinking and ability to solve the problem.
Interestingly, some good engineers fail to be great because they are also associated with their code. The AI ​​-infield code forces them to make critical estimates and improve them, rather than occupying their work. This change can lead to better quality of code and less recovery issues.
In 2011, Mark Anderson said, “The software is eating the world.” I am sure this is still true – more than ever before. AI and software are becoming necessary in every human endeavor.
When we enter a personal period, every individual and every business – regardless of the size – will find solutions according to their specific needs. AI tools will unlock a large wave of empowered software engineers through automation and intelligence.
We are entering an interesting time for product creators, service providers and software engineers. AI will not replace engineers. This will increase their capabilities. The future is for people who embrace these tools to make better, faster and faster solutions.
