Resume | LinkedIn | Stack Overflow
Hello, I am Kabir. I am a Real World data Engineer at the Medicines and Healthcare Products Regulatory Agency (MHRA).
Previously, I completed the Digital, Data and Technology Graduate scheme at NHS England as a Data Engineer. I also worked as a Data Engineer at Health Data Research UK (HDRUK). I hold an MSc in Artificial Intelligence from the Queen Mary University of London and completed my BSc in Software Engineering.
Description: Developed deep learning based drone detection system that able to detect any flying drone in real time. I have used You only look once (YOLOv4) object detection framework to implement this project. I have trained 6000 imgaes (4000 was manually labelled by labelimg tools) to traine this model and tested in live environment.
Description
VGGNET16: Developed a flower classifier using VGG16. Other tools I used keras and some machine learning concept such as Activation, MaxPooling2D, BatchNormalization, Flattening, EarlyStopping etc.
ResNET20: Implemented CIFAR10 classifier using ResNet20 including machine leanring tools/ technology such as TensorFlow, optimizers, BatchNormalization, AveragePooling2D, ZeroPadding2D etc.
Description: Developed basic hand digit recognition classifier and trained on MNIST data.
Description: This is NLP project divided into 4 parts includes Word Embeddings with Word2Vec, Basic Text Classification, Using LSTMs for Text Classification and Text classification using BERT.
Neural Machine Translation: An NMT system which translates texts from Spanish to English using a Bidirectional LSTM encoder for the source sentence and a Unidirectional LSTM Decoder with multiplicative attention for the target sentence (GitHub).
Dependency Parsing: A Neural Transition-Based Dependency Parsing system with one-layer MLP (GitHub).