I am working at KPIT Engineering on CARIAD projects for the Volkswagen Group through Technica Engineering GmbH.Client: Volkswagen GroupProject: BAP(Bedien und Anzeigeprotokoll) ProjectResponsibilities:Analyzing and reporting comprehensive validation results.Acting as the main contact for customer interactions and resolving urgent requests.Delivering detailed test reports, including defect tracking and quality metrics for seamless client review.Collaborating with cross-functional teams and working closely with developers.Utilizing Scrum methodology in project execution.Creating KPM and TC ID tickets for project tracking.Key Technologies and Tools:BAPViewer, Wireshark, CAN, DoIP,TCP/IP,Ethernet, Network Management, Diagnostics, XML, Python, Visual Studio Code, Excel, Jira, GIT, GitLab, Scrum.

Web application to detect anomalies of electrocardiogram from ECG images with artificiel intelligence.-Annotation ECG images with Roboflow to detect ECG signals with YOLOv8 for verfication the input in application is an ECG image or not .-Classify cardiovsculaires diseaes from 12-leads of ECG images using YOLOv8 for classification,CNN2D and algorithme of machine learning like:SVM,XGBOOST,KNN.-Implement the best model "CNN2D" with a flask web application.Key words:ECG ,cardiovsculaires diseaes,12-leads,YOLOv8,CNN2D,SVM,KNN,XGboost,Flask

Analysis of Iris and Pupil Constriction and Dilation with AIData Collection: I collect datasets by recording videos of individuals' eyes.Data Preprocessing: The recorded videos are processed by framing them to extract individual eye frames.Annotation: I annotate these eye frames into 2 classes (iris and pupil) using the Roboflow platform, creating an instance segmentation project. This results in a dataset with annotations.Dataset Export: After annotation, the dataset is exported and prepared for model training.Model Selection: I employ the YOLOv8 ultralytics model, specifically YOLOv8x-seg, for iris and pupil segmentation.Model Training: The model is trained on the annotated dataset to learn the segmentation task.Performance Evaluation: We evaluate the model's performance using various metrics to ensure accurate iris and pupil detection.Testing: The trained model is tested on both images and videos to validate its effectiveness.Application Development: We develop a user-friendly application using Streamlit. This application allows users to upload images or videos for iris and pupil detection.Output: For images, the application provides detected iris and pupil regions with bounding box values. For videos, also the application provided detected videos ,it displays a plot showing changes in pupil size over time.Key Words:Deep Learning,Computer Vision,Segmentation,YOLOv8,CNN,Streamlit

Detection of tomato plant leaf diseases with deep learning and deployment of the model with a mobile application. (Mention: Very Good)-Architectures(CNN)used:AlexNet,ResNet152V2,ResNet50,VGG16,InceptionV3,DensNet121,MobilNet-IDE: Kaggle, Google Colab, Android Studio-Frameworks:Tensorflow,Tensorflow Lite,Pytorch-Libraries: Keras , Numpy , Matplotlib,....

I did two tasks of Spark Foundation -Task 1: Prediction Using Supervised ML-Task2:Prediction Using Unsupervised ML

Recognition of human activity-Collect Dataset with Smartphone and put them in Edge Impluse platform- Train the model on the Edge Impluse platform-Train the model with coding on google colab and using the CNN architecturekey words :Human Activity Recognition,Edge Impluse,Deep learning,CNN