Abdulhusein Jabir Poonawala

• Developed web-based software for remote AGV control using React, MongoDB, and JavaScript; containerized on AWS EC2 with CI/CD pipelines. Used Material UI & TypeScript.
• Automated unit testing with GitLab CI/CD for both positive and negative cases.
• Integrated new VMS features like filtering and highlighting using Odoo Programming on Linux, following Agile methodology.

Conducted computer simulation and laboratory material specimen impact testing to discover novel material combinations.

Deployed 3D Modelling Algorithms and their User Interfaces through LUA and XML scripting.

• Employed Machine Learning libraries (Keras - TensorFlow, Darknet-YOLO, Caffe) to make and train 15+ smart city models with 93% overall accuracy.
• Composed a lightweight & efficient, illegal swimmer and boat intrusion detection using python OpenCV.
• Compiled data points via MySQL querying for visual display through Microsoft Power BI.

• Experienced in building ML pipelines using Scikit-learn, with focus on feature engineering, hyperparameter tuning, and metrics like RMSE, AUC-ROC.
• Skilled in deploying ML solutions with Docker and Kubernetes, ensuring scalability, availability, and security.
• Proficient in automating CI/CD pipelines for ML models using GitHub Actions, MLflow, and Kubeflow.

Grade: First Class

Activities and societies: Dean's List for AY18/29 and AY20/21 Sem2. Participated in various workshops and projects related to engineering and technology.

First Class | Highest Distinction
Specialisation in Aeronautical Engg
Major in Design and Innovation
Minor in Computer Science
NUS Overseas College (Munich)
CAP: 4.71/5.0
AY18/29 & AY20/21 Sem 2 - Dean’s List

• Applied statistical techniques like Linear Regression, Naïve Bayes, and Decision Trees with regularization methods to prevent overfitting and improve model accuracy.
• Built and optimized deep learning models, including CNNs, LSTMs, and GANs, for tasks such as feature extraction, time-series analysis, and synthetic data generation.

• Implemented KNN for traffic sign classification with dataset preprocessing, feature extraction (HoG, LBP), and model evaluation.
• Learnt in object detection (YOLO, SSD), image segmentation, flood fill algorithms, K-means clustering, and CNNs with Keras.
• Applied video processing like background subtraction and object tracking (Mean-shift, Camshift) via OpenCV.