I am an aspiring researcher in Statistics and Data Science, with a strong interest in applying robust statistical methods and machine learning to solve complex, real-world problems.

My core research interests include:

• 🤖 Bayesian Inference & Decision Theory: Building models that quantify uncertainty and drive optimal, data-driven decisions.
• 📈 Time-Series Forecasting: Developing predictive models for supply chain and demand planning, especially in dynamic environments.
• 🔊 Signal Processing & ML: Extracting meaningful features from complex signals (like audio) for classification tasks.
• 📡 IoT & Sensor Fusion: Architecting end-to-end systems for collecting and analyzing real-world data for monitoring and control.

This section highlights the projects associated with my academic papers. Each repository contains fully documented and reproducible code.

• What is in a Price? Estimating Willingness-to-Pay with Bayesian Hierarchical Models

  • An application of Bayesian Hierarchical Conjoint Analysis to deconstruct product price into the monetary value of its features.
  • Accepted at the 2025 Advances in Science and Engineering Technology International Conferences (ASET).
  • [GitHub Code] | [Paper - arXiv Link]

• DemandLens: Enhancing Forecast Accuracy Through Product-Specific Hyperparameter Optimization

  • A Prophet-based time-series model incorporating external regressors (COVID-19 data) for SKU-level demand forecasting.
  • Accepted at the 2025 Advances in Science and Engineering Technology International Conferences (ASET).
  • [GitHub Code] | [Paper - arXiv Link]

• Profit over Proxies: A Scalable Bayesian Decision Framework for Optimizing Multi-Variant Online Experiments

  • A novel A/B/n testing framework that aligns statistical analysis directly with business profitability using Bayesian decision theory.
  • [GitHub Code] | [The paper is currently under review.)

• Machine Learning Framework for Audio-Based Equipment Condition Monitoring

  • A systematic and statistically rigorous comparison of ML algorithms for industrial fault detection from audio signals.
  • Accepted at the 2025 Advances in Science and Engineering Technology International Conferences (ASET).
  • [GitHub Code] | [Paper - arXiv Link]

• CattleSense - A Multisensory Approach to Optimize Cattle Well-Being

  • An end-to-end IoT system using Arduino and Raspberry Pi for real-time monitoring of livestock health and environmental conditions.
  • Published in the 2024 Advances in Science and Engineering Technology International Conferences (ASET).
  • [GitHub Code] | [Paper - Official DOI Link]

• 🔭 I’m currently working on... Refining the "Profit over Proxies" research project for journal submission, and exploring scalable MLOps pipelines.
• 🌱 I’m currently learning... Deeper aspects of Probabilistic Machine Learning (PML) and building robust data systems on cloud platforms like AWS and Azure.
• 💬 Ask me about... Python, Bayesian Inference, Time-Series Forecasting, and building end-to-end reliable, robust, and scalable data pipelines.

• Languages: Python, SQL, MATLAB, R, JavaScript, C++ (for Arduino)
• Data Science & ML: Scikit-learn, Pandas, NumPy, PyMC, Prophet, TensorFlow, XGBoost
• Data Visualization: Matplotlib, Seaborn, ArviZ
• IoT & Hardware: Arduino, Raspberry Pi, PlatformIO, ESP32 Microcontroller
• Tools & Platforms: Git, GitHub, Jupyter Notebooks, Docker

• LinkedIn: https://www.linkedin.com/in/srijeshpillai/
• Google Scholar: [https://scholar.google.com/citations?view_op=list_works&hl=en&hl=en&user=Y-aaRxwAAAAJ]
• Email: [[email protected]]