Waqas Ahmed

WAQAS AHMED Data Scientist | Machine Learning Engineer | Analytics Consultant Islamabad, Pakistan Professional Summary Data Scientist and Machine Learning Engineer experienced in designing end-to-end, realworld analytics and machine learning systems with measurable business impact. Skilled in forecasting, predictive modeling, NLP, and decision-support analytics, with a strong focus on translating data into actionable insights. Professional Experience Data Science Intern — AI-GenMat (Jun 2025 – Aug 2025) - Integrated Retrieval-Augmented Generation (RAG) pipelines improving contextual accuracy of AI workflows - Reduced manual data preprocessing time by ~30% through reusable Python pipelines - Assisted in evaluating ML and neural network models, improving experiment turnaround time Projects Formula 1 Race Outcome Prediction System • Engineered 25+ race features from multi-session data • Achieved MAE ≈ 2.4 race positions, outperforming baseline heuristics by ~35% • Enabled scenario analysis for race strategy simulations Enterprise Sales Forecasting & Demand Intelligence Platform • Improved forecast accuracy by ~20% over naive baselines • Enabled inventory planning simulations reducing stockout risk • Delivered executive-ready forecasting dashboards Customer Churn & Lifetime Value Prediction Engine • Identified high-risk customers with ~75% precision • Enabled targeted retention strategies reducing churn risk by ~15% • Produced actionable segmentation insights Automated Analytics & ML Platform • Reduced analysis turnaround time by ~60% • Enabled non-technical users to generate insights autonomously • Automated EDA, modeling, and reporting workflows Financial Risk & Credit Scoring System • Improved risk classification accuracy by ~18% • Enabled explainable risk scoring for decision transparency • Reduced false approvals in simulated lending scenarios Core Skills • Data Science Core: Python (Advanced), Pandas, NumPy, SciPy, Statsmodels, Jupyter Notebook, Data Cleaning Pipelines, Feature Engineering Workflows, Statistical Modeling • Machine Learning: Scikit-learn, XGBoost, LightGBM, CatBoost, Logistic Regression, Random Forests, Gradient Boosting, K-Means, Hierarchical Clustering, PCA, GridSearchCV, Optuna • Deep Learning: TensorFlow, Keras, PyTorch, CNN Architectures, Transfer Learning, Custom Training Loops, Model Checkpointing, Early Stopping • Natural Language Processing (NLP): NLTK, spaCy, Hugging Face Transformers, SentenceTransformers, TF-IDF, Word2Vec, BERT-based Models, Text Classification Pipelines, Sentiment Analysis Systems • Large Language Models & AI Systems: OpenAI API, Hugging Face Inference API, Retrieval-Augmented Generation (RAG), FAISS, ChromaDB, Vector Embeddings, Prompt Engineering, Model Context Protocol (MCP) • Data Visualization & Business Intelligence: Power BI (Advanced), DAX, Power Query, Tableau (Basic), Matplotlib, Seaborn, Plotly, Interactive Dashboard Design • Data Engineering & Pipelines: SQL (Advanced), MySQL, PostgreSQL, MongoDB, ETL Pipelines, Data Ingestion Scripts, Schema Design, Query Optimization • Deployment & MLOps (Applied): Streamlit, FastAPI, Flask, Docker, GitHub Actions, Model Serialization (Pickle, Joblib), RESTbased Inference APIs • DevOps & Tooling: Git, GitHub, Linux CLI, Bash, Virtual Environments, Conda, Poetry • Cloud & Platforms: AWS (EC2, S3 – Working Knowledge), Google Colab, Kaggle, Vercel (App Hosting) • Languages: Python, SQL, JavaScript, C++, C, Java, Assembly (MASM615)