We Generated a 20-Slide Investor Deck With Speaker Notes in 3 Minutes

The Problem

Investor decks are a tax on founders.

You need one for every fundraise, every board meeting, every LP update, every pitch competition. Each one takes 8-15 hours if you're doing it right. That's a designer in Figma or a founder fighting PowerPoint at 2 AM, manually adjusting text boxes and pretending they know what "good slide design" means.

And every deck is 80% the same content reshuffled. Problem slide. Solution slide. Market size. Team. Traction. Ask. You've built this deck before. You'll build it again next quarter with updated numbers.

At PyratzLabs, we create decks constantly — for portfolio companies, for internal presentations, for events. We needed a way to go from "here are the key points" to "here's your deck, it's done" without the 8-hour design tax.

The Solution

An Mr.Chief agent that takes a markdown brief, generates a 20-slide PPTX using python-pptx, applies consistent branding, creates data visualizations, and writes speaker notes for every slide.

Feed it your bullet points. Get back a deck you can present tomorrow.

The Process (with code/config snippets)

The input is a structured markdown brief:

# Deck Brief: PyratzLabs Q1 2026 LP Update

## Branding
- Primary: #4F46E5 (indigo)
- Secondary: #10B981 (green)
- Font: Inter
- Logo: ./assets/pyratzlabs-logo.png

## Slides
1. Title: Q1 2026 LP Update | subtitle: PyratzLabs Portfolio Review
2. Agenda: Portfolio overview, Key metrics, Notable exits, Outlook
3. Portfolio Summary: 12 active companies, 3 new investments, 1 exit
4. Key Metrics (chart: bar):
   - AUM: $42M → $58M
   - Deployed: $31M
   - Reserved: $12M
   - Dry powder: $15M
5. Fund Performance (chart: line):
   - Q1'25: 1.2x | Q2'25: 1.4x | Q3'25: 1.3x | Q4'25: 1.6x | Q1'26: 1.8x
...
20. Closing: Thank you | contact: bilal@pyratzlabs.com

The agent processes this through a python-pptx pipeline:

from pptx import Presentation
from pptx.util import Inches, Pt, Emu
from pptx.dml.color import RGBColor
from pptx.chart.data import CategoryChartData
from pptx.enum.chart import XL_CHART_TYPE

prs = Presentation()
prs.slide_width = Inches(13.333)  # 16:9
prs.slide_height = Inches(7.5)

# Slide master with branded colors
slide_master = prs.slide_masters[0]
for layout in slide_master.slide_layouts:
    for ph in layout.placeholders:
        ph.text_frame.paragraphs[0].font.name = 'Inter'

# Title slide
slide = prs.slides.add_slide(prs.slide_layouts[0])
slide.shapes.title.text = "Q1 2026 LP Update"
slide.placeholders[1].text = "PyratzLabs Portfolio Review"

# Chart slide
slide = prs.slides.add_slide(prs.slide_layouts[5])
chart_data = CategoryChartData()
chart_data.categories = ['AUM', 'Deployed', 'Reserved', 'Dry Powder']
chart_data.add_series('$M', (58, 31, 12, 15))
chart = slide.shapes.add_chart(
    XL_CHART_TYPE.COLUMN_CLUSTERED,
    Inches(1), Inches(1.5), Inches(11), Inches(5),
    chart_data
).chart
chart.series[0].format.fill.solid()
chart.series[0].format.fill.fore_color.rgb = RGBColor(0x4F, 0x46, 0xE5)

# Speaker notes for every slide
slide.notes_slide.notes_text_frame.text = (
    "Key callout: AUM grew 38% QoQ driven by portfolio markups "
    "and new LP commitments. Dry powder positions us well for "
    "2-3 follow-on rounds in Q2."
)

prs.save('Q1-2026-LP-Update.pptx')

Every slide gets speaker notes. Not generic filler — contextual talking points based on the data in the brief.

The Results

The quarterly update is the killer use case. Same structure, new numbers. The agent regenerates the entire deck with updated data in 3 minutes. No "find-and-replace across 20 slides." No "did I update the chart on slide 14?"

Try It Yourself

Write your deck as a markdown brief. Be specific about branding (colors, fonts, logo path). Include data for charts as structured lists.

The agent handles layout, formatting, charts, and speaker notes. You handle the content and the room.

A founder's time is worth $500-1,000/hour. Spending 15 hours on a PowerPoint is burning $7,500-15,000 of founder time. Spend 20 minutes writing a brief instead.

The best investor decks aren't designed — they're structured. Structure is exactly what agents are good at.