Uses

A comprehensive list of tools, libraries, frameworks, and platforms I use daily for machine learning, AI development, and data science work.

ML & AI Frameworks

PyTorch is my go-to deep learning framework for building and training neural networks — I use it for NLP models, classification tasks, and custom architectures.
TensorFlow / Keras for rapid prototyping and deploying production-ready models, especially for image and text classification pipelines.
Scikit-learn is my standard toolkit for classical ML — random forests, SVMs, logistic regression, clustering, and preprocessing pipelines.
XGBoost for gradient boosting on tabular data, particularly useful in financial risk forecasting and disease prediction models.
HuggingFace Transformers for fine-tuning pre-trained language models on domain-specific NLP tasks such as summarisation and sentiment analysis.

Pandas and NumPy form the backbone of every data pipeline I build — from cleaning and transforming raw datasets to feature engineering at scale.
Matplotlib and Seaborn for exploratory data analysis, statistical plots, and presenting insights in research papers and reports.
Plotly and Streamlit for building interactive dashboards and deploying data apps quickly — my career aspirations predictor is live on Streamlit Cloud.
Power BI and Tableau for business intelligence reporting and creating executive-level dashboards from structured datasets.

VS Code is my primary editor. I rely heavily on the Python, Pylance, Jupyter, and GitLens extensions for day-to-day ML development.
Jupyter Notebooks for iterative experimentation, EDA, and sharing reproducible research — every ML project starts in a notebook before moving to a module structure.
FastAPI is my preferred framework for wrapping ML models into production REST APIs — fast, async-friendly, and auto-generates OpenAPI docs.
Flask and Django for lightweight APIs and full-stack web apps respectively, including the document summariser backend.
Git and GitHub for version control, collaborative development, and CI/CD pipelines across all projects.

Computer Vision with OpenCV, YOLO, and CNN-based architectures for object detection, image segmentation, and super-resolution — building on the deep learning work from my third internship.
Retrieval-Augmented Generation (RAG) systems that pair large language models with vector stores (FAISS, Pinecone) to build context-aware, knowledge-grounded AI applications.
LLM fine-tuning using HuggingFace PEFT / LoRA to adapt foundation models to domain-specific tasks efficiently and at low cost.
Advanced Deep Learning — Transformers, diffusion models, and multi-modal architectures — through hands-on projects and the fast.ai and DeepLearning.AI curriculum.
MLOps practices: experiment tracking with MLflow, containerisation with Docker, and CI/CD pipelines to move models from notebook to production reliably and reproducibly.

Primary Language	Python
Other Languages	SQL, Java, JavaScript, React, Node.js
Relational Databases	MySQL, PostgreSQL, SQLite
NoSQL	MongoDB
Cloud	AWS, Azure (fundamentals)
Education	BSc Computer Science — University of Sindh (Dec 2025)
Experience	3x Intern (ML, Data Science, AI/DL) at ITSOLERA PVT LTD