Fast facts: Data analytics and insights

Q What’s the most fascinating task for data analytics and insights specialists?

Working with machine learning and artificial intelligence (AI), including using the latest small task-specific language models and computer vision models to navigate lots of unstructured text and discovering trends from satellite imagery.

Q What’s a new trend in data analytics and insights that’s exciting?

Data and analytics engineering techniques and platforms that support democratisation of data and insights for our clients.

Q Career vs reality: what’s working in data analytics and insights really like?

There’s a misguided perception that working with data effectively translates to sitting at a computer – either coding or building dashboards. Yet, working in the field of data and insights requires a ‘people person’, and it’s much more of a relationship role. 

Essentially, this role is about building trust to help solve complex problems. It demands the ability to listen deeply to clients, understand their unique business requirements and pain points, and help to build successful solutions. Every project is effectively a new problem to solve, so it is the ideal career for someone who loves solving problems. At the end of the day, having a client or internal stakeholder state, “wow, how did you do that?” is a reward.

Working in data and analytics is about being pragmatic, innovative, and nimble – there is never a dull moment! In fact, the RPS data analytics and insights team have had a gamut of exciting opportunities arising, including developing AI solutions for the Clean Energy Regulator, and supporting the Department of Transport and Planning to develop a new plan for Victoria.

While our team work directly with clients across a range of great projects, we also collaborate between our other specialist divisions and our parent company Tetra Tech – maximising the benefit of our expertise. We operate as a community with a small core team with other data scientists and software engineers supporting on a project-by-project basis across Australia and internationally.

A typical day consists of…
After pouring ourselves a large cup of coffee, usually, our day starts with catching up on emails. Most likely you have a client meeting coming up in a few days’ time, so after you’ve perused your inbox, you ensure your client data pipeline is working and seamless. Your job is to serve up current, high-quality, and clean data to the clients Power BI dashboard. 

Next, you move onto the Power BI dashboard. Sifting through all the pages, you put yourself in your clients’ shoes to ensure the data presentation is visually appealing – aiming for features that are intuitive and seamless. You might refer to the set of requirements and design agreed with the client to add new features to the dashboard.

This pipeline contains un-structured open-text data, so we use our qualitative data processing platform AutoThematic to add themes to the open-text data to be integrated in the client’s dashboard report. Rinse and repeat.

What’s the greatest challenge in data analytics and Insights?
The biggest challenge is the fast pace at which the industry is evolving. Tools and technology in data engineering, data science, and AI/ML are rapidly updated and new technologies appear almost monthly. It’s important though that while we remain at the forefront of change and innovation for our clients that we don’t get caught up in the hype-cycle.

Q What’s your favourite resource for data analytics and insights news?

‘The Batch’ by Andrew Ng is a newsletter that’s usually a fountain of information. Social media such as LinkedIn and Reddit are great for tracking the latest trends in tech.

Q Who’s a living legend in the field of data analytics and insights?

Tom Aarsen is a machine learning engineer at Hugging Face – which is a platform where the machine leaning community collaborate on models, datasets, and applications. This is the team that are driving the open-source AI revolution. Their movement makes AI open, accessible, and safer for all. Tom is a prolific contributor to key packages that data scientists rely on for processing large amounts of text. He posts and shares updates on LinkedIn regularly to keep the community in the loop.