Hi! I'm Raha Moosavi, you can call me Raha

A graduate data scientist with one years of machine learning industry experience in the area of deep learning and recommendation systems. Working on a range of classification and optimisation problems. Ability to deliver innovative solutions accurately and efficiently solve challenging business problems. Nimble and agile professional to deliver solutions in a rapidly changing environment. Curious and inquisitive with a constant desire to learn and pioneer new uses for machine learning.

Currently doing my Internship with TotalEnergies

Working as machine learning scientist in TotalEnergies. Particularly, working on creating deep learning surrogate model for reservoir simulation.

During my time in Imperial College London

Student expected to graduate with Second Upper Honors in Computational Science. Focusing on machine learning, data structure, algorithms, programming and software developement. I also am doing my internship in TotalEnergies as a machine learning scientist for Machine Learning and data science department. My project is about building deep learning based model for reservoir simulation, using attention recurrent residual U-net and a dataset from geological map and pressure and saturation map as output.

participated in 6 hackthons and group projects

Our team got the 1st place (out of 15 teams) to build an accurate deep learning model for classification of lung CT Scan images to Covid-19 detection. This machine learning project has 3 main parts, gathering and preprocessing data (using image augmentation), building model (on Pytorch), and building a user-interface for working with model(on Grio). There are many challenges, from the data preprocessing such as unbalancing dataset, data augmentaion, to the limitation on choosing model and increasing test data accuracy. Integrating every component and working in a team together within 2 weeks.

Teaching

Driverless car-Applied Machine learning

  • identify and define data-oriented problems and data-driven decisions in real life
  • discuss and illustrate the problems in terms of image exploration and visualization
  • apply basic machine learning tools to extract inferential information from the data
  • learn how to apply Computer Vision and Deep Learning techniques to build automotive-related algorithms
  • learn how to use essential Computer Vision techniques to identify lane lines on a road

My role

  • prepare tutorial
  • share from my experiences on how I would approach real-world problems
  • share cool and interesting external materials and from my writings to spark their interests in data science and artificial intelligence
  • students approach me to discuss on data science problems outside course work

Content covered

  • Basic of Python and Numpy
  • Basic of percepton
  • Computer vision
  • Deep Neural Network
  • Multiclass classification
  • Data Dashboards