Day in the Life of a Computer Programmer

Professional Spotlight: Sunil Kowlgi

Why did you decide to pursue computer programming?

I was introduced to computer programming in a class in sixth grade in the 1990s. The class was based on the Logo programming language, in which you write commands to draw things on a computer screen. I loved programming in Logo as I could create any shape or figure from my imagination.

For the final project, I created an alien robot, and although the teacher only asked for the program, I also submitted a short story on the alien robot. The teacher gave me a good score on the computer program and gave me bonus points for the story. From that initial experience, it stuck with me that programming was a fun activity and a cool way to express your imagination.

I later took programming classes in high school and college, but I wasn't yet thinking of a career in computer programming. It was during a project in my senior year of college that I realized I enjoyed programming and was good at it. This project was different in that it was entirely self-directed — I came up with the project proposal and wrote 1,000-1,500 lines of code by myself, and the project worked pretty well. The project was to emulate a software protocol called CAN, which is used for inter-device communication in automobiles.

For example, your car will sound an alarm when a door is open because a sensor in the door sends a CAN message to the car computer. After the project, I was sure that I wanted to have a programming job, and that's what I've done ever since.

What are the biggest challenges you have faced working in computer programming?

The biggest challenges I've encountered are related to complex code bases that are co-written by many software programmers. An example of a complex code base would be software code for the Chrome browser, which goes into the hundreds of thousands, if not millions, of lines of code.

When working on complex software, you rarely ever write code from scratch. Even when a company decides to make a brand new product, you start by reusing existing software code and modifying the code to meet product requirements. To modify code, you first need to comprehend it.

There are two challenges when it comes to comprehending an existing code base. The first challenge is that people writing software are focused on making the product work while making the code comprehensible is usually an afterthought.

But as time goes by and the software gets reused by others not familiar with the code, the comprehensibility of code starts to matter. If code is hard to understand, software programmers new to the code might end up introducing bugs. This failure to properly understand code before reusing it is one of the most common reasons for software bugs.

When working with a complex code base, you need documentation that describes the code architecture and how the code works. A challenge in most cases is the documentation is either stale, as it hasn't been updated as the software code is updated, or doesn't have all the information. When you don't have good documentation, it slows down your ability to understand and use the code correctly.

What are the most rewarding aspects?

When I see my software products help someone be successful in whatever they do, that is the most rewarding feeling. Users are making videos using my software for teaching others about cryptocurrency, the Spanish language, and math problem-solving skills, among others, and these videos have reached tens of thousands of their subscribers. It feels awesome to help users deliver great video content to their audience.

Was it challenging to find a job in the field?

When I graduated from college in the 2000s, there were plenty of software jobs available, and that is still the case even today. But it was a bit of a challenge to find jobs in a particular niche that I preferred, which had to do with writing device driver software (software that interfaces with electronics hardware).

I was lucky to come across an ad for a company called Kyocera Wireless that had openings for mobile phone software development and landed a job there.

What did your career trajectory look like after earning your degree?

My first software job was at Kyocera Wireless as an entry-level software programmer, where I wrote software in C for mobile phones. This was a time before the iPhone came into the market. I worked there for two years and then took a break to get a master's in computer engineering at the University of Texas at Austin.

After my master's, I worked seven years as a software programmer at National Instruments in Austin. There, I worked on device driver software in C++ for microprocessors and circuit boards. The best part of working at National Instruments was being in the company of really smart programmers, from whom I learned a lot. That environment helped me develop the skills and confidence to do more ambitious things, like building a software startup.

For the last four years, I've been working on startup projects in the video space. I built Android apps, web apps, and server software in Java and Javascript. Today, I have a software product called Outklip that is used by more than 2,000 people (and growing) to make tutorials, demos, and gameplay videos for YouTube.

How do you organize, plan and prioritize your work?

Coding is a lot like playing a game. You do some amount of planning upfront, then write some code and see the results. Then plan some more, write more code, and so on. It's a gradual process where you have to complete a level to get to the next level, and there are no cheat codes. You must develop code in steps, or it won't end up working right.

Before writing any code, I work things out on paper or a whiteboard. Writing things down helps clarify my thinking. It's on paper where I come up with an outline of what I want to accomplish.

I draw diagrams to represent data, code execution sequences, and code architecture. Once I have figured out how I'm going to tackle a problem, I start coding the easiest parts first. I write the bare minimum code to generate a tangible result — like a basic webpage that I can see and interact with — and then I repeat cycles of plan and code through to completion.

Advice for newcomers to the profession?

Programming is a skill you learn by doing. Work on projects. They're the best way to learn. Textbooks are helpful after you've done some coding already; they're best for refining your understanding and improving your knowledge about programming.

To improve your skills, look for examples of code online, tweak them, and run to see results. Keep modifying code till you get the results you want. If you repeat this cycle of starting with examples and ending up with something cool, you're on your way to becoming a very effective programmer.