Aaron Chew

I start my mornings with breakfast and a bit of surfing on social media to keep up with the news, mostly on emerging technologies. As an Artificial Intelligence (AI) engineer in DSTA’s Digital Hub, I explore opportunities in emerging technologies that can be harnessed for defence. In particular, my interest is in finding new ways for machines to understand the human language so that they can be used to help the Singapore Armed Forces (SAF) in their operations. Hence, it is important to be aware of the latest developments and releases.

In our work, we collaborate with commercial organisations and academia, where we hold online video meetings or trainings with them. My team works closely with a global AI firm on a co-development project to make AI technologies simpler to implement. We have regular discussions as part of our agile development process, to share our progress as well as to quickly address and resolve any issues the team faces. Internal discussions are also held within the team or with our SAF colleagues to brainstorm ideas and tackle challenges together. On top of that, we review trial results against our hypotheses from the previous day, and plan new steps to improve the AI models.

I spend some time before lunch to check emails or browse through new AI developments, conferences, or research papers. It is interesting to look up new ways or explore new open-source tools to better approach some of the problems we face in our projects. My colleagues and I also regularly attend classes, training and conferences to beef up our AI knowledge since the field is rapidly expanding.

I send some outstanding experiments to run in the background and write down some potential leads I can look at after lunch.

If I am in the office, I usually lunch with my colleagues at our office cafeteria. As we work on a lot of interesting technologies, our lunch conversations typically turn into discussions on new technology developments, like the latest tech companies, new releases of software and hardware or even ideas on how we can improve the way we work. These conversations can sometimes develop into ground-up initiatives which we may pitch to bring to fruition, as our workplace culture supports innovation at all levels!

I usually spend the afternoon running experiments and tackling my to-do list and leads. This usually takes a few hours, or until I achieve a milestone troubleshooting bugs or scripting for a new task. Once done, I send my analysis for the team to review and extend upon. I will also schedule new experiments to run overnight.

In 2020, I also had the opportunity to help organise BrainHack, DSTA’s learning fest for students to experience and learn about emerging technologies. In particular, I helped out with the Today I Learned – AI Camp, where we introduced AI and machine learning concepts to participants. It was a refreshing change from my usual job scope, and also a very fulfilling experience to share our knowledge with students and inspire their passion for tech.

I typically end the day with an informal discussion with my team members where we wrap up our day sharing new findings and helping one another brainstorm for new ideas and solutions.