Skip to content
View aliamarliana's full-sized avatar
:dependabot:
:dependabot:

Block or report aliamarliana

Block user

Prevent this user from interacting with your repositories and sending you notifications. Learn more about blocking users.

You must be logged in to block users.

Maximum 250 characters. Please don't include any personal information such as legal names or email addresses. Markdown supported. This note will be visible to only you.
Report abuse

Contact GitHub support about this user’s behavior. Learn more about reporting abuse.

Report abuse
aliamarliana/README.md

👋 Hi there! I'm Alia Marliana

🎓 Master of Data Science (USM, CGPA 3.88) | BSc Software Engineering (UNITEN, CGPA 3.98)

💡 I specialize in turning raw data into actionable ML systems. I’m a big believer in rigorous data preparation and genuinely enjoy the challenge of cleaning complex datasets to build better predictive models, NLP tools, and RAG pipelines. I bridge the gap between messy data and production-ready code to deliver results that drive real business value.

🧠 Core Skills

  • Programming: Python · R · SQL · Java
  • ML/AI: scikit-learn · TensorFlow · XGBoost · CatBoost · NLP (spaCy) · LLMs & Generative AI · RAG Pipelines · Model Evaluation & Hyperparameter Tuning · Computer Vision
  • Data & Visualization: Pandas · NumPy · Matplotlib · Seaborn · Power BI · Tableau · Streamlit · Figma
  • Databases & Tools: MySQL · MongoDB · Cassandra · ChromaDB · Git · Jupyter Notebook · Google Colab · FastAPI · LangChain · HuggingFace

📝 Selected Projects

(Note: some projects are not in my personal repository, so I’ve linked directly to the original team repos or design files.)

  1. Role-Based RAG Chatbot for FinTech – Production-ready chatbot with FastAPI + Streamlit, HuggingFace LLM, and ChromaDB for secure, role-based document retrieval.
  2. Rossmann Sales Forecasting – Time series and ML models for micro/macro sales forecasting (SARIMA, XGBoost, CatBoost).
  3. Term Deposit Subscription Classification – Binary classification models with feature selection and hyperparameter tuning.
  4. Health Disease Prediction – Predictive modeling for heart disease using classification algorithms, with robust feature selection and class imbalance handling.
  5. R Coffee Co. Dashboard – Power BI dashboard built on POS-style sales data to analyze menu performance, peak demand periods and customer spending patterns, supporting data-driven decisions on staffing and promotions.
  6. ZUS Coffee A/B Testing Experiment – Conducted an A/B test on promotional strategies in the ZUS Coffee mobile app. Compared emotion-driven limited-edition merchandise (Variation A) vs price-focused combo discounts (Variation B) to evaluate impact on conversion rate and user engagement. Findings demonstrated higher conversions and engagement for emotion-led campaigns.

🚀 What I Do Best

  • Build end-to-end machine learning solutions from data ingestion to deployment-ready workflows
  • Translate analytical results into clear insights and dashboards for decision-making
  • Develop production-ready ML applications using Python, APIs and lightweight interfaces
  • Work comfortably across modeling, data analysis and system integration

📫 Connect

LinkedIn | Explore my AI/ML projects on GitHub

Pinned Loading

  1. finbot-rbac finbot-rbac Public

    Forked from codebasics/ds-rpc-01

    A full RAG-based Role-Based Access Control (RBAC) chatbot using FastAPI, LangChain, ChromaDB, and HuggingFace models — built for the Codebasics GenAI Resume Challenge.

    Python

  2. rossmann-sales-forecasting rossmann-sales-forecasting Public

    A sales forecasting project using SARIMA, CatBoost, and XGBoost models to predict daily revenue for Rossmann retail stores, comparing traditional time-series and modern machine learning approaches.

    Jupyter Notebook

  3. product-defect-classification product-defect-classification Public

    An object detection pipeline using TensorFlow and KerasCV (RetinaNet) to identify manufacturing defects in jar lids. Features custom data processing and handling of ragged tensors for object detect…

    Jupyter Notebook

  4. cafe-sales-powerbi-dashboard cafe-sales-powerbi-dashboard Public

    A business intelligence dashboard built in Power BI to transform R Coffee Co.’s sales data into actionable insights for menu optimization, staffing efficiency, and customer engagement.

  5. steam-games-recommender-part2 steam-games-recommender-part2 Public

    A performance analysis of multiple recommender system models — User-to-User CF, Item-to-Item CF, and Content-Based Filtering — to evaluate accuracy, ranking metrics, and optimization strategies for…

    Jupyter Notebook

  6. zus-coffee-ab-testing zus-coffee-ab-testing Public

    Data-driven A/B testing case study analyzing promotional strategies on the ZUS Coffee mobile app using product analytics principles.