I've been working in the areas of data science and machine learning for many years.
In my latest projects, I performed end-to-end DataOps and MLOps, including deployment to production, on decarbonization potential of housing in the UK. I also developed and productionized several highly accurate machine learning models for energy consumption and efficiency of UK homes.
Some of other projects I did are a) Machine learning on natural products (chemical compounds), which is applying different supervised and unsupervised machine learning tasks on chemical molecules data, and b) unCoVer: Unravelling Data for Rapid Evidence-Based Response to COVID-19, which is again on applying machine learning tasks on the COVID data.
For the above-mentioned projects, I've been using several tools and libraries including, but not limited to, Python, Spark, Azure, MLflow, Docker, Jupyter, TensorFlow (Keras), scikit-learn, pandas, NumPy, XGBoost, Git (GitHub), Azure DevOps, pytest, Bash and Matplotlib.
I have developed several softwares, of which the sources of a few are publicly available at my GitHub account. Note that my most recent projects have been private, so the sources are not publicly shared in the GitHub account, but I'm open to provide an overview of what was my experience doing them.