Statistician · Autodidact · Bibliophile

About Me

I’m a Data scientist and Statistician with a focus on applying statistical thinking to real product and business decisions. My interests and expertise lie at the intersection of applied statistics, machine learning, and open-source development — turning rigorous analytical ideas into practical tools, software, and workflows ¹.

Open Source Projects

Please feel free to clone/fork projects, raise issues and submit PRs if you think something could be better.

🧩 Shafayet Khan Shafee
├─📦 R Packages
│ ├─ MOR                   - Calculate Median Odds Ratio From a Fitted Multilevel Binary Logistic Model
│ ├─ skmisc                - package with some useful functions to make life easier
│ ├─ chunkthis             - Rstudio Addin to create code chunks with copied R code
│ └─ hugodownplus          - Extend 'hugodown's markdown document with toc and html boxes
├─🐍 Python Packages
│ ├─ skmiscpy              - A python package providing Contains a few functions useful for data-analysis, causal inference etc.
│ └─ kedrogen              - A simple CLI command that generate a Kedro project from a cookiecutter template in the current directory.
├─📜 Bash Scripts
│ └─ uvshot                - Python setup scripts using uv
├─🐳 Docker Images
│ ├─ mad-coc-IPW           - Contains docker image to reproduce of an analysis results
│ └─ contained-thesis      - Contains docker image to reproduce of a simulation results
├─📊 DataViz
│ ├─ WlidLife Strike       - A dashboard showing plane damage by wildlife strike
│ └─ TidyTuesday           - My contributions to the #TidyTuesday weekly data project
├─🔵 Quarto Extensions
│ ├─ reveal-header         - Add header text and header logo in all pages of RevealJs slide
│ ├─ line-highlight        - Implement code line highlighting for HTML docs
│ ├─ downloadthis          - Embed download buttons in HTML docs
│ ├─ interactive-sql       - Run sql examples interactively in RevealJs slide or HTML files
│ ├─ collapse-callout      - Make the Callouts collapsible globally in HTML docs
│ ├─ metropolis-beamer     - Another Quarto reveal.js template extension inspired by Beamer Metropolis theme
│ ├─ metropolis            - Quarto reveal.js template extension inspired by Metropolis theme
│ ├─ bsicons               - use Bootstrap Icons in HTML docs and RevealJs presentations
│ ├─ nameref               - Use named reference in Quarto Docs
│ ├─ add-code-files        - Add contents from external source or code files
│ ├─ code-fullscreen       - Adds a full screen button in the code blocks in Quarto docs.
│ ├─ collapse-social-embeds- Make collapsible callouts with Github gist, tweets, youtube vids etc
│ ├─ black-formatter       - Format the codes in python code chunk using black
│ ├─ material-icons        - Use Google Material Design Icons in Quarto docs
│ ├─ hide-comment          - Add comments in code chunk, hidden in rendered output
│ └─ style-speaker-note    - Style speaker notes of the RevealJs slides from a css file
└─💡 Other
  ├─ DBT Demo Project      - A Demo DBT Project
  └─ Beamer Templates      - My collection of beamer templates

Happy Coding! 😊

Skills

• Stats & ML: Causal Inference · Experimentation · Multilevel Modeling · Bayesian Inference · Time Series Forecasting · Predictive Modeling · Statistical Testing
• DataOps & MLOps: DBT · Kedro · MLflow · Docker · VM · Bash Scripting
• Dev: R · Python · SQL · Lua
• Other: Rmarkdown · Quarto Extensions · LaTeX