Karan Pathak

– Led the development of a novel transpiler that transpiles code from sparql to cypher for neo4j and benchmark comparison with other state of the art strategies available. – Developed a neo4j extension based on the investigative results of query optimisation strategies of graph databases.

Contributions: - Developed novel neural net architectures involving Graph Neural Networks with a combination of other Deep learning models for prediction of drug sensitivity in cancer which was benchmarked with state of the art models. - Integrated and encoded various data types as graphs for learning graph representations and interpreted the results of neural network with the help of hypothesis testing.

Involved in a research project with algorithms for the detection of multivariate outliers in PV module data sets. Tasks included: - Planning the procedure - Data analysis and filtering - Implementation of routines of linear algebra/machine learning in Python - Documentation, visualization and communication of research results - Literature research

- Spearheaded multiple projects with variable team sizes of up to 5 people. - Provided AI enabled scalable solution to forecast the sales for all departments in different stores of a retail client. - Developed recommender systems based on collaborative filtering for a mood based application. - Responsible for creating ETL pipelines, performing feature selection, machine learning, evaluation, extensive back-end scripting and research. - Prototyping, wire-framing, white-boarding ideas, creating RESTful API's

- Specialization in Artificial Intelligence – Involved in a collaborative thesis and research publication titled “Integrating Multi-Omics Data and Pathway Structure with Explainable Graph Neural Networks for Precision Medicine”. – Awarded 2nd position for Freicar self driving car project that involved computer vision, deep learning and mobile robotics.

This MicroMasters program in Statistics and Data Science is comprised of four online courses and a virtually proctored exam Courses include: - Probability - The science of uncertainity and data - Fundamentals of statistics - Machine learning with python: from Linear models to deep learning - Data Analysis: Statistical Modeling and Computation in Applications - Capstone exam in Statistics and data science