“The analysis is only as good as the data that comes in”
1. Could you share your professional journey and what led you to your current role as a Business Analyst at HCI Australia?
I’ve always been interested in working with data and analytics. My interest in analytics began when I was introduced to Formula 1. It’s the pinnacle of motorsport where every second matters. How are these decisions made? Everything is controlled by algorithms and data-driven decisions. I pursued a master’s degree in analytics at RMIT because I wanted to improve my skills in this area. During my final semester, I interned with another healthcare provider organization, where I worked on real client data and completed predictive analytics projects. This was an eye-opening experience to learn how data can be put to use. Around that time, the emergence of Generative AI also began. When ChatGPT was launched, I dived right into the heart of the system, learning how it operates and performs. My study led to many breakthroughs, and I also created some side projects that put the technology into action. The first application I created was for the education industry, where I fine-tuned one of the OpenAI models that would primarily assist students in grades 8 through 10 in preparing for exams. It was able to present students with a unique and innovative type of learning experience in which they could literally chat to their syllabus, take mock examinations, develop mind maps, and clarify any questions they had. This was my introduction to the world of Large Language Models (LLMs). When I was looking for opportunities, I came across HCI’s business analyst vacancies and realised that my experience and expertise in AI and analytics could be very useful to the organisation. HCI Australia was the place where I anticipated putting my skills to the test while also learning a lot about the education industry.
2. What drew you towards working with data, analytics, and technology in the healthcare and sustainability space?
Formula 1 was the key reason I became interested in data analytics. Once the motivation was established, I learnt a variety of principles that could be applied to real-world events with significant impact. What pulled me into the healthcare and sustainability field was seeing the power of AI and how it might be used to make a difference in the business. AI is a very strong tool that may be used to automate a variety of tasks, saving countless hours in any sector. Applications include automated resource allocation, intelligent irrigation systems, and many more.
3. How have your past academic or professional experiences shaped your interest in AI and predictive analytics?
I received my Masters Degree from RMIT with a focus on data analytics. I was always curious about how to play with data, so I wanted to learn more about the fundamental nature and math that goes into all algorithms and how they function. Throughout my studies, I learned about how neural networks, regression- based algorithms, and other prediction algorithms function. This laid the groundwork and equipped me with the ability to create something unique. I began researching the applications of LLMs and how they can be implemented in real-world systems. I worked on a number of projects that included a variety of prediction algorithms, ranging from complex ones like deep neural networks to simpler, but more effective ones such as decision trees. Who knew the world of probability could be so interesting? Among the many projects I had completed, the most notable ones were projecting which automobile models would be most suited for the Indian car market among all models and manufactures in the globe. This involved considering a variety of aspects such as the car’s size and weight, transmission, cost, fuel efficiency, and more. My contemporaries at the time praised the algorithm’s accuracy.
4. What do you find most rewarding about working at the intersection of business analysis, technology, and healthcare sustainability?
The most intriguing component of this would be to see how things from different perspectives come together. Initially, I had no concept how much technology is engaged in the sustainability and healthcare industries. The more I understand about how things work and how data analysis principles may be applied to this, the more intrigued I get, which fuels my motivation to continue working at the intersection of these. And this is the most enjoyable aspect of my work because each day is unique. When it comes to problem solving, finding bottlenecks in existing business processes, and devising solutions to them, no two days are alike. It truly keeps you on your toes. Furthermore, this space is experiencing rapid evolution on its side. Keeping up with all the changes in technology and its applications is challenging today, but the knowledge gained will be the most gratifying aspect.
5. From your perspective, how can AI and machine learning be applied to predict environmental risks in healthcare systems? Could you share an example from your work or observation?
Artificial intelligence and machine learning have already proven to be quite useful in forecasting environmental concerns and healthcare systems. A couple of examples I came across included the technologies used in tornado and earthquake-prone areas. The initial piece of this jigsaw would be ultra-sophisticated sensors that collect all raw data for the computers to forecast. Once that is done, complex algorithms such as random forest can be utilized to accurately anticipate the outcomes. From using ML and AI to identify environmental dangers, there are numerous applications that have already been implemented and are saving countless lives in the case of a natural disaster. To put it simply, the Gradient Boosted Trees algorithm can forecast tornado likelihood and classify tornado magnitude. This approach is already being used in several experimental projects that produce excellent results, based on historical data and additional model fine tuning. Another example is the usage of Convolutional Neural Networks, or CNNs, which combine multiple deep learning algorithms to create predictions based on sensor data and previous data. When people talk about AI now, they usually refer to the generative AI boom, which is simply asking your computer to do a variety of chores for you based on a simple prompt and use cases created around it. However, artificial intelligence goes beyond that. They are usually business-facing or utilized for highly particular and specialist purposes, which is why many individuals are unaware of them. The previously described CNN algorithm serves as the foundation for artificial intelligence. Using AI, environmental risk prediction and mitigation measures might possibly be automated. Many applications, such as the Roomba robot and various Amazon warehouses, employ AI to automate jobs so that they can be completed as efficiently as possible, saving firms countless hours in maintenance and operating procedures.
6. Have you worked on projects where predictive analytics helped optimize resources (like energy, water, or clinical supplies) while still maintaining efficiency?
While I was studying, I had an internship that allowed me to really hone my predictive analytics skills. The project’s goal was to anticipate when a specific portion of a gas turbine will break based on the oil’s state and a variety of other sensor data. This was for a lubrication company that pushed for predictive maintenance of gas turbines based on sensor data and oil quality. This meant that their vendors’ costs for maintenance could be reduced significantly, and the quality of the oil could be monitored in real time. It was an extremely thrilling project.
7. In your experience, how do healthcare organizations respond to adopting AI/ML tools for environmental or resource optimization—are they eager, cautious, or resistant?
The potential applications of AI and machine learning in healthcare are highly interesting. With the recent surge in AI- powered applications, there is a strong desire among industry executives to push for as widespread AI adoption as feasible. All thanks to the promise that AI will produce remarkable results. Based on my own experience, people are more likely to adopt ML-based solutions than AI-based solutions. This is simply because ML approaches are more predictable and consistently produce similar results when applied to applications. AI, on the other hand, can be quite valuable in taking the fundamentals of machine learning and building automation on top of proven algorithms to save even more time and effort. Unfortunately, as previously said, when people talk about AI these days, they mainly mean generative AI, which promises to be the jack of all trades, but does not always produce consistent results. It is incredibly difficult to identify edge cases and implement suitable safeguards to avoid repeating the same mistakes in the future. Generative AI, and how LLMs function in general, is analogous to a superb probability algorithm. What it does is simple: it predicts the next most probable character or word. Some people call it magic since the interactions appear to be human-like. However, it makes many blunders, as do all humans. The reality that all LLMs and how they work are essentially black boxes, and no one can really predict what will happen if the same questions or inquiries are asked repeatedly. To summarize, when it comes to real-world AI and ML use cases, people are often excited because they perceive potential. Applying generative AI, on the other hand, has many people on the fence, and they are typically a mix of enthusiastic and resistive until an MVP has been built.
8. What role do you see real-time data and IoT sensors playing in strengthening predictive models for environmental sustainability?
IoT devices play a critical role in improving predictive modelling applications for environmental sustainability. When real-time data from a variety of sensors, such as humidity, temperature, and CO2 levels, is collected, reliable predictions of harsh weather and other natural disasters can be made. There are currently numerous apps that leverage raw data from these IoT-style sensors to help make real-time choices. However, the analysis is only as good as the data that comes in. It is critical to ensure that IoT sensors are of excellent quality for accurate predictive modelling and the associated use cases.
9. Have you seen examples where AI-driven insights not only reduced environmental risks but also cut costs or improved operational resilience?
Yes! Take Tesla and Amazon, for example. They make heavy use of artificial intelligence to create automated gigafactories that run nearly entirely on their own, with little to no human interaction. Algorithms excel at repeated tasks, as seen by their numerous uses. Tesla’s gigafactory saves hundreds of hours and reduces operating costs when producing newer Tesla vehicles. This allowed them to produce cars at a far faster and more precise rate than the average manufacturing worker. From a sustainability aspect, there are examples of handling enormous volumes and requirements for data centres while optimizing their resources. Smart farming is also gaining popularity, slowly but surely, and it is pretty exciting. If smart farming is used as an example, it can significantly reduce environmental risks while streamlining the entire operational aspect by saving time, optimizing resources, and assisting in producing high-quality crops that benefit everyone.
10. Based on your journey so far, what message would you share with healthcare leaders and institutions about the value of AI and predictive analytics in building a sustainable and resource-conscious future?
If I had to convey a message to healthcare leaders, it would be to adopt AI and ML-driven use cases as early as possible. The curve has shifted; if you lack the necessary talents or are unable to quickly adjust to this new reality, you will fall behind. AI and predictive analytics, in general, hold immense potential for the future; invest in the correct areas and reap the long-term rewards.