jupdeck 0.1.0

Create Audience-Ready Presentations from Jupyter Notebooks

📘 JupDeck

Convert Jupyter notebooks to Presentation-Ready Slide Decks.

Ideal for:

• Technical consultants
• Data scientists
• Instructors
• Teams presenting notebook work to non-technical stakeholders

---

What It Does

This tool transforms .ipynb notebooks into structured .pptx presentations.

• Markdown Headers become slide titles
• Markdown Bullets become slide bullets.
• Plots and Tables become slide graphics.
• Markdown Text becomes speaker notes.

---

Quick Start

Pip install the package: pip install jupdeck

Convert a notebook to slides:

jupdeck convert examples/data/jupdeck_overview.ipyng outputs/output.pptx

---

JupDeck Roadmap

This roadmap outlines the phased development plan for the JupDeck tool. The goal is to build a system that transforms Jupyter Notebooks into professional PowerPoint presentations, supporting data science reporting, executive summaries, and technical documentation.

1: Baseline Slide Generator

Goal: Prove core functionality: parse notebooks and render working PowerPoint output with embedded graphics.

Features:

• Parse notebook markdown structure and code cell outputs
• Generate PowerPoint slide decks (.pptx)
• Embed output graphics (e.g., plots, images, tables) into slides with generic/simple formatting
• PyPI and GitHub releases with versioning

Notes: "User in the loop" for clean up of formated slides necessary

---

2: Notebook Slide Directives

Goal: Interpret slide layout directives from Jupyter notebooks

Features:

• [] Rudimentary directives can be embedded as html-comments in markdown cells
• [] Directives control simple formatting, e.g. "New Slide", "Hide Tables"

---

3: Improved Layouts

Goal: Automatically Adapt Slide Layouts

Features:

• [] 3 different content slide layouts offered plus intro / ending layouts
• [] Layout rules encoded in reusable config file
• [] During slide rendering, best layout for slide chosen based on contents
• [] Format validation during rendering (e.g. image resolution, text overflow)

---

4: Custom Theming and Templates

Goal: Use Custom Styling Templates and Themes

Features:

• [] Choose between 3 custom styling templates (e.g. dark mode, corporate, academic)
• [] Add user-defined templates

---

5: Custom Content Filtering

Goal: Generate different decks depending on audience

Features:

• [] Directives mark content for different rendering scenarios (e.g. technical vs. non-technical, internal vs. external)
• [] Command-line flag controls rendering scenario and generates appropriate deck (e.g. jupdeck input.ipynb --theme internal --no-code)
• [] Reusable configuration profiles (YAML)

---

6: VS Code Plugin

Goal: Notebook rendering from VS Code

• [] Plugin allows notebook rendering from VS Code interface
• [] Plugin provides suggestions during notebook editing

---

7: Basic LLM Editing Support

Goal: Use LLMs for editing support Features:

• [] Summarize markdown into concise, audience-appropriate bullets
• [] Align slide tone with target audience (e.g. "executive-friendly", "technical peer review")

---

8: LLM Slide Enhancement

• [] Integrate conversational interface to notebook
• [] Auto-suggest slide titles and bullets from raw code/markdown
• [] Validate presentation claims against content: "This model seems to be overfit", "This chart does not support the title claim"

---

9: Team Collaboration & Workflow Integration

Goal: Support real-world enterprise workflows

• [] GitHub Actions to auto-render slides from notebooks on PR merge
• [] Enable shared template libraries across teams
• [] Add comments/annotations for reviewer workflows

💡 Why?

Jupyter notebooks are fantastic for exploratory work — but hard to present clearly. This tool turns notebook narratives into shareable reports for clients, collaborators, and classrooms.

---

🧪 Development & Testing

poetry run pytest         # Run test suite
poetry run pre-commit run --all-files  # Run linters and formatters

---

📄 License

Apache License 2.0 — permissive for personal, educational, and commercial use. (See LICENSE file)