Hi! I am Ahnaf Asif Khan. I am a senior software engineer at Enosis Solutions Ltd, Bangladesh. I have received my Bachelor of Science in Electrical and Electronic Engineering from Islamic University of Technology (IUT), Bangladesh, in 2021. My research interest primarily revolves around Cyber-Physical Systems and Internet of Things, a domain in which I have also been working on as a software engineer.
I am a sports enthusiast, I like to jog, or play whenever I get chance, and I watch a lot of soccer. In my off times, I read books, watch movies and TVs, sitcoms are always my go-to genre.
Aims to develop techniques to detect botnets during initial scanning and propagation phases with ensemble machine learning models.
Aims to fortify models by training them with adversarial examples to anticipate and counteract evasion tactics.
With a modified K-Means clustering algorithm to factor in Manhattan distance, the model proposed cluster medians as HUBs.
A distribution network for rechargable batteries of Electric vehicles, through clustered HUBs, from central Solar Farms to Customers.
Using CARLA simulator, lane detection and navigation of an autonomous vehicle in the context of Bangladesh through a PID controller.
Implementation of the YOLOv5 object detection algorithm integrated with a Super-Resolution GAN model to enhance the image resolution.
A Generative Adversarial Network model to enhance resolutions of 320x240 pixels images by 4 times, where separate Generator and Discriminator networks are combined with GAN Loss and Adam Optimizer. [Code]
A neural network model to provide solar radiance prediction, trained on a dataset provided by the HI-SEAS weather station. The model was evaluated by plotting loss against mean squared error, explained variance and coefficient of determination, and yielded a satisfactory outcome. [Code]
A multi-layer network implemented with CISCO Packet Tracer to simulate enterprise level network integration, configured with advanced protocols including OSPF, EIGRP, STP, and HSRP to ensure efficient routing, with secure WAN connections utilizing Frame Relay and PPP, alongside ACLs and port security to bolster network defenses. [Code]
A voice-command-enabled image editor with implementations of OpenCV and pre-trained Neural Style Transfer models. The application used Neural Style Transfer models alongside matrix manipulation to provide multiple options for image editing, through voice command. [Code]
A Five-course specialization covering Convolutional Neural Networks and Deep Learning, Improving Deep Neural Networks with Hyperparameter tuning, Regularization and Optimization, Structuring Machine Learning Projects, Convulational Neural Networks, and Sequence Models. [Certificates]
A Four-course specialization covering Natural Language Processing with Classification and Vector Spaces, Probabilistic Models, Sequence Models, and Attention Models. [Certificates]
Three courses covering self-driving software stacks, vehicle modelling, control and safety frameworks, State estimation and localization through sensors, visual perception including visual odometry, object detection and tracking. [Certificates]
Three courses covering Python functionality, features, and Query DataFrame structures used for data science, Applied Plotting, Charting & Data Representation in Python, and Applied Machine Learning in Python. [Certificates]