- Worked on projects like Predicting Malware Classification and Family using machine learning models with automated feature selection to improve detection accuracy and speed- Utilised Python libraries like Scikit-learn, Pandas, and NumPy for feature engineering and model optimisation in malware classification projects- Developed Memory-Augmented Deep Recurrent Neural Networks (MDRNNs) to address long-term dependency issues in natural language processing tasks, improving model performance in text analysis- Implemented deep learning models using TensorFlow and Keras to build a Fully Residual Convolutional Neural Network for brain tumour segmentation and classification across diverse medical imaging modalities- Gained hands-on experience with data cleaning, feature engineering, and visualization tools like Matplotlib and Seaborn for clear communication of project findings- Used Jupyter Notebooks to document my work and conduct exploratory data analysis, iterating through datasets to uncover valuable insights that informed model development. This tool also allowed me to present findings in a visual and easy-to-understand manner for my team.

- Implemented Convolutional Neural Networks (CNNs) and Recurrent Neural Networks (RNNs) for image and sequence data analysis.- Optimised neural network architectures by tuning hyper-parameters (learning rate, batch size, etc.) to improve model accuracy and performance.- Used Python libraries like Pandas, NumPy, and Scikit-learn for data preprocessing and model evaluation.- Developed and trained neural networks using TensorFlow and Keras to solve classification and regression problems.- Conducted model evaluation using metrics such as accuracy, precision, recall, and F1-score, to ensure high-quality predictions.- Applied deep learning techniques to projects involving image classification, text analysis, and recommendation systems.

- Explored advanced Python libraries and tools, such as BeautifulSoup for web scraping, asyncio for asynchronous programming, and pickle for data serialization.- Enhanced understanding of Python libraries such as NumPy, Pandas, and Matplotlib, enabling efficient data manipulation and visualization.- Implemented Convolutional Neural Networks (CNNs) and Recurrent Neural Networks (RNNs) for image and sequence data analysis.- Conducted model evaluation using metrics such as accuracy, precision, recall, and F1-score, to ensure high-quality predictions.- Implemented advanced error handling techniques, using try-except blocks, the raise keyword, build robust and error-resistant code.

- Worked on the development of personalised recommendation engines to improve customer engagement and business strategies for enterprise clients.- Built and optimised Natural Language Processing (NLP) models for text classification, sentiment analysis, and topic extraction from large datasets- Pre-processed large-scale unstructured textual data using techniques like tokenization, stemming, lemmatization, and vectorization to ensure high-quality input for machine learning models.- Implemented algorithms like transformers and recurrent neural networks (RNNs) using the library TensorFlow.- Applied feature engineering and dimensionality reduction techniques (PCA, t-SNE) for model optimisation, and utilised Python libraries (Pandas, NumPy, Matplotlib) for data cleaning, analysis, and visualization to communicate insights effectively.

- Conducted research on microbial biofilms, focusing on their formation and resistance in oral environments.- Synthesised exopolysaccharides (EPS) from Bacillus amyloliquefaciens (B2SA2) to create an oral biofilm for antimicrobial testing.- Developed an EPS-based nano-composite with AgNO3 and SDS, designed to inhibit bacterial adhesion and biofilm formation on dental surfaces.- Utilised typodont models to simulate oral conditions and test the efficacy of nanocomposite coatings against oral pathogens.- Applied advanced techniques like UV-VIS spectrophotometry and crystal violet staining for analysing biofilm formation and antimicrobial efficacy.- Analysed results from scanning electron microscopy (SEM) to confirm the structural properties of biofilms and the success of the nano-composite coating.

- Explored the intersection of biology and data science acquiring foundational knowledge, focusing on computational analysis of biological data.- Gained hands-on experience in bioinformatics tools and software to analyse DNA sequences and protein structures.- Developed and implemented data preprocessing pipelines for cleaning and preparing biological data for further analysis.- Worked on a project involving the alignment and annotation of DNA sequences to identify gene variants linked to specific traits or diseases.- Applied computational techniques to manipulate and analyse large datasets, particularly focusing on protein structure prediction using bioinformatics algorithms.

- Conducted data exploration and visualization to derive insights and ensure data quality prior to model input, enhancing the overall analysis process.- Developed and deployed machine learning models using Python, TensorFlow, and scikit-learn, achieving notable predictive accuracy through effective model architecture design and training.- Utilised PyCaret's AutoML for model training and selection, deploying the application with Flask for real-time predictions and integrating it into a cloud environment for scalable performance. - Analysed patterns and trends in the data and performed feature engineering to enhance model performance, utilising Pandas and NumPy for thorough data preprocessing and cleaning to address missing values.

- Conducted research in Reinforcement Learning, Distributed Machine Learning, Unsupervised Learning, and Text Data Mining.- Led extensive literature reviews to analyse and summarise key insights in Machine Learning, Deep Learning, Big Data, and Distributed Computing. - Collaborated with academic and industry clients to publish high-impact research papers, adhering to top-tier journal quality standards.- Designed and executed experiments for AI-based speech signal segmentation using Kernelized Deep Networks.- Delivered comprehensive data-driven reports and contributed to the advancement of AI research, ensuring all research outputs met or exceeded both client expectations and scientific benchmarks.- Generated and proofread project-based papers with precise results, ensuring plagiarism-free content and adherence to journal guidelines, while assisting researchers in writing high-quality papers and addressing review comments to enhance their research profile.