Read an email. Send a text. Search the Internet. Play a video game. You’ve probably done one or more of these things today at the tap of a button. Yet under the surface, each one of them relies on software—complex documents with thousands of lines of computer code, written and tested by human hands, that tell a device what to do and how to do it.
Just as our everyday digital apps are built on code, so are many experimental projects in labs around the world. While researchers are often the face of new discoveries in computer science, their work is also often built on the efforts of many others. These others include not just fellow scientists, but engineers who help translate ideas on the page into working prototypes for testing and improvement.
At A*STAR, the Software Engineering Community of Practice (CoP) was launched in May 2022 to bolster the recognition and identity of software engineering professionals within the agency’s research community. The CoP also aims to facilitate knowledge sharing, catalyse professional development and establish a supportive network for its members at A*STAR.
In part one of a feature series in collaboration with the CoP, A*STAR Research speaks to Su Myat Phyoe and Jia Yi Loo, software engineers from A*STAR’s Singapore Institute of Manufacturing Technology (SIMTech) and the Institute for Infocomm Research (I2R) respectively, to learn more about the nuances of their profession, their motivations and how their work supports research successes.
Q: What brought you to A*STAR?
SMP: From prior experience in the research industry, I became aware of A*STAR’s impressive reputation in R&D. I joined the agency as I was drawn by the prospect of facing new challenges and embracing continuous learning, particularly in the areas of machine learning and artificial intelligence (AI) for manufacturing.
The environment here perfectly aligns with my career aspirations, and I’m enthusiastic about the opportunities A*STAR offers to grow both professionally and personally.
JYL: I applied to A*STAR after graduation since its cybersecurity department and research base caught my eye. Cybersecurity was a hot topic then, and still is. It’s a key part of software engineering, because it’s ultimately about how to code in a way that keeps systems safe.
I was interested in the field since I’d worked on Android malware analysis and detection during my final year project (FYP) in university. Funnily enough, I didn’t have much programming experience before my FYP, which taught me the basics.
Q: What do you do as software engineers?
SMP: I work with scientists to create custom software tools tailored to their needs, such as managing and analysing research data efficiently, automating repetitive tasks, integrating different technologies, and providing user-friendly interfaces.
Tools like these can make research work more accessible, efficient and productive, which enables scientists to focus on their research endeavours and achieve meaningful results.
Q: How do you think software engineers help enable research successes?
SMP: The contributions of software engineers in research enhance efficiency, accuracy, collaboration and communication, all of which promote more successful and impactful research outcomes. Their expertise supports researchers in addressing real-world problems and making significant advancements in their fields of study.
JYL: I believe we play a very supportive role to scientists. Building on insights or an overview of the kind of research a scientist wants to do, software engineers will try to translate the scientist's ideas into prototypes they can test, or later on, products that can be used by the public.
Q: What motivates your work?
SMP: I love being a software engineer in the research field because it lets me learn something new each day. At SIMTech’s Cyber-Physical Production Systems (CPPS) group, we aim to develop smart decision-making technologies for the manufacturing industry. Projects like predictive maintenance play a crucial role in boosting the industry’s efficiency and leading the way to Industry 4.0.
I also love having the chance to share the knowledge I’ve acquired with industry professionals by conducting training courses and collaborating on projects with them. It lets me contribute to their growth and make a positive impact in their industry. My passion for continuous learning and teaching drives me to excel in my work and keeps me excited about future challenges and opportunities.
JYL: With coding, you can see the entire process behind the product and how everything works together. You can always keep playing around with code. The best part comes when you see that code come to life as its end users run it, and know that you contributed to making it happen.
To be candid, I enjoy the opportunities to build things that can help people in their day-to-day life; I am motivated when I hear that what we built are things that people use, and that they make people happy.
With the rise of new cybersecurity threats, I also hope what I do helps protect the software systems around us, whether it’s through supporting the research or creating the products involved.
Q: What are some interesting projects you’ve worked on?
SMP: One of my recent projects at SIMTech involved developing predictive maintenance models, which was particularly challenging due to limited data resources. However, it was incredibly rewarding to overcome those obstacles and utilise natural language processing techniques to extract valuable insights from machine records.
At the project's conclusion, we successfully developed a web-based application that delivers real-time machine failure prediction results using live data.
We also applied microservice technology in that project as it improves scalability, allowing us to handle growing demands efficiently and adapt swiftly to changes in the manufacturing industry.
JYL: At I2R, I work mostly in malware, moving beyond Android to Windows systems. My work usually supports research by government agencies. In one project with the Centre for Strategic Infocomm Technologies (CSIT), we looked at changes in antivirus labels and malware classifications over time—how a file can sometimes fluctuate between being recognisable malware and a benign sample.
I’m currently working on a project that aims to automate factchecking, which is a bit different from dealing with malware. Today, factchecking is almost always done by journalists who have to sift through a lot of articles to verify a claim. We’re trying to see if large language models can assist us in that process and detect the spread of fake news.
Q: What advice would you give a future software engineer?
JYL: One thing that will give you an edge is to really explore the technology and build up technical skills where you can. The tech moves very fast, so you’ve got to keep in touch with what’s available right now.
The most important thing is to be adaptive. At any time, you may have to work on products or software that’s completely new to you, but don’t stress too hard about it; just do your best and learn as you go!
