Conversations with Teo Shu Qi

By Lin Min Htoo and Claudia Tan

This article was produced in collaboration with the Maritime Singapore Connect (MSC) Office, a unit under the Singapore Maritime Foundation (SMF). 

Teo Shu Qi is a Data Scientist at Alpha Ori Technologies, a maritime technology company operating in IoT (Internet of Things), Ship ERP (Enterprise Resource Planning) and Big Data science. A Data Science and Analytics graduate from the National University of Singapore, Shu Qi is involved in the digitalisation of maritime operations and using data and machine learning techniques to help in the decision making during operations.

At Alpha Ori Technologies, we come up with real-time data-driven decisions that help ships and shore crews operate smoothly. I joined Alpha Ori in May 2020 and have since worked on many interesting projects. To me, the role as a maritime Data Scientist is unique compared to other industries which already have technologies and AI in place as the maritime industry is on its technology advancement journey and there are many areas I can learn from.

I remember applying through LinkedIn as well as the government career portal (Careers@gov). I read a lot on the job descriptions to decide which roles I was interested in before actually applying, especially looking at the technical aspects of the role. I wanted to make sure the application of Data Science was impactful and interesting. If you are looking for an internship or career opportunities in the maritime industry, you can also visit the MSC website ( for more information.

There had been quite a few challenges as I joined with limited knowledge on maritime. In fact, everyone from my team came from non-maritime background! I remember researching on the maritime industry in order to devise Data Science applications that could make processes more effective and efficient. I also approached my colleagues who were keen to teach me what they know. While it was a huge learning curve, it was also what made my job exciting.

A game changer has been achieved by connecting ship to shore through real-time monitoring system of ships. This involves installing of servers and hardware on the ships so that we can get regular data about the current conditions from the ship, even if it is far away in the middle of the Pacific Ocean. For example, one aspect we are monitoring is the fuel consumption of the ships. As fuel is a major cost in shipping (60% of the ship’s operational costs), many maritime companies are looking to minimise its usage for commercial reasons and also due to the increased focus on decarbonisation and sustainability. Using data science models, we can analyse factors such as futuristic weather conditions, distance to go, real-time ship’s operational data such as speed, geolocation coupled with the ship’s performance modelling to determine the optimal speed for the ship from time to time, with the objective of enabling the ship to meet the expected date & time of arrival with the least amount of fuel consumed.

Firstly, there are many opportunities to make a difference as the industry is seeking new technology ideas to help the entire ecosystem. Secondly, the issues I get to engage with are very unique, making every day in this line of work different and a whole new experience. For instance, in situations similar to the Suez Canal blockage, technology and data science may possibly prevent the incident from happening. 

In my case, one of the systems we had to improve was the manual planning of loading and unloading of cargo in oil tankers from ship to shore. We reviewed the historical records and existing processes and designed a system where we can leverage AI-based technology to automate the planning process so that it is safer, maximises the stowage and reduces cargo claims.

Understanding maritime operations and its processes were the biggest challenges as I had limited knowledge of the industry’s inner workings when I first joined. Hence, it took me some time to understand how the systems operated. Moreover, there was quite a bit of engineering and physics involved which I struggled with.

Once the pandemic situation normalises, I look forward to visit a ship calling Singapore to understand first-hand how the systems are interconnected and critical for ships’ reliable operations.

Communication is vital as we need to liaise with project managers and clients to help them translate technical data we observed into business insights for them. Having a very good data model is not very useful unless the client or the business can understand how to utilise the model.

One way is to observe people by watching how they approach the problems and how they communicate with others. This allowed me to build the soft skill needed by Data Scientists.

The modules I took back in University definitely gave me a good understanding of statistics, mathematics and the different data science methods for use in my job. Hence, I was well equipped with the knowledge needed to do my job and could apply what I learned in the day-to-day operations.

At first, I was not set on a particular industry when I was studying Data Science. My interest in the maritime industry sparked from a module I took back in school on Logistics and Transportation. The module, conducted as a series of hackathons, introduced me to create solutions to real-world situations and data contributed by maritime companies, which were not the typical problems we faced before in school. This motivated me to join the maritime industry as the problems were new and challenging which got me interested and curious about this industry.

You should take online courses or university modules in mathematics and statistics as this knowledge will help you to better understand methods in machine learning and deep learning. This knowledge will also help you to adapt and learn in the future.

Maritime is an age-old industry and very critical for the world’s economy – as evidenced when the container ship ‘Ever Given’ choked the Suez Canal. The potential in maritime is huge as AI-based applications are in the nascent stage of development. There is very good growth potential for aspiring young data scientists. There is no dearth of pain points waiting for innovative solutions, here.

I wish that I could have chosen modules based on what was really useful for my career rather than modules that I could score well in just to maintain a good CAP/GPA.

There are plenty of courses and materials online on Data Science on Coursera and YouTube. Learners can tap on these resources to learn key concepts and build a solid foundation. They can also consider working on Data Science projects or joining competitions on platforms such as Kaggle to gain more experience.

Data Science is heavily used in industries like e-commerce, but in other industries like maritime, there are many opportunities to increase the adoption of Data Science. It does take some persuasion and encouragement to overcome the inertia from the stakeholders in these industries to introduce something new and different. Data scientists also need to gain a good domain understanding of the problems in that specific field, which takes some time and effort, before they can create useful solutions.

Learning how Data Science is really used and applied in the real world was very eye-opening. In school, one can learn many theoretical algorithms and methods. However, learning how to apply those models in practice is most valuable to me and keeps me excited to go to work every day.

I hope to continue doing Data Science for the next 5-10 years. Perhaps, I will take up a Master’s to further hone my skills and learn more.

Getting some work experience before deciding which Master’s programmes to enrol in could help to narrow down which fields are of interest to you, what you really want to learn and what you really like to do.