Build end-to-end machine learning (NLP) solutions, deep learning models based on neural networks for multilingual texts using Python Spark, and improve model performance by over 20%. Create a human-in-the-loop machine learning system which lowered the business cost of algorithmic errors by 70%. Deploy machine learning models into production in serverless on AWS and set up MLOPs pipelines for all models. Design and conduct A/B experiments for various versions of the company’s APIs; write unit and integration tests. Develop APIs using Flask framework for rapid prototyping of ML proof of concepts for stakeholders.

Collaborated with a team of 4+ data professionals and engineers on the cleaning, analysis and modelling of energy consumption/monitoring data using Microsoft Excel and Python; leading to a reduction in overhead. Handled technical documentation for engineering projects.