Science or engineering?

Often data science and data engineering will get confused. Data engineering is the design and implementation of databases, pipelines and flows to make your data more accessible: Merging datasets, aggregating results and setting up live feeds to help process complex data structures are all part of the data engineering process.

Data science involves analysing and interpreting data to create new insight: identifying underlying relationships, predicting future behaviour, constructing new measures to understand complex data.

Brainnwave has expertise in both of these areas, but it’s good to know the difference!

The data science flow

1

Exploration

At this point our data scientists work with you to understand and prepare the data for the right model. We analyse the data holistically, as well as keeping the current use-case in mind.

2

Prototype

We consider a suite of models that can provide a potential solution to your problem. We use a variety of performance metrics to determine which model is most suitable and then fine-tune selected models for the best results.

3

Production

This is the point where we select the best model and make it operational, working with you to ensure the results are easily interpreted and intuitive.

Data Science

Your data science team

Got a team? Already working on data science internally? Great! We can work alongside and support your team to provide our insight and reach the best solution. Many hands make light work.

No internal team? No problem. We can provide the knowledge and expertise required to handle the data science for your company. We’re happy to run the data science projects from scratch as required.

Data science in practice...

Gas flaring solutions

We used satellite data to identify flaring locations from heat signatures so our client could target flaring locations with their green energy solutions.

Market resilience index

We built a global market resilience index to help our client understand which drinks markets to invest in accounting for things like strength and variablility.

ASB community profiles

We produced a suite of machine learning models to help housing officers understand anti-social behaviour in their area in order to better service tenants.

Solar power prediction

We used large amounts of IoT data from a solar panel array to predict future power output and identify ineffective panels and aid maintenance.