---
title: "Reconstructed Slides: Arrakis: How To Build An AI Sandbox From Scratch - Abhishek Bhardwaj, OpenAI"
category: "slides"
video_id: "wsFd22SL1s8"
sourceLabels: ["Cropped public YouTube video frames", "Local OpenCV slide-region detection", "Local RapidOCR"]
---

# Reconstructed Slides: Arrakis: How To Build An AI Sandbox From Scratch - Abhishek Bhardwaj, OpenAI

## Source Video
[Arrakis: How To Build An AI Sandbox From Scratch - Abhishek Bhardwaj, OpenAI](https://www.youtube.com/watch?v=wsFd22SL1s8)

## Method
This deck is reconstructed from the existing video frame captures by detecting likely slide regions with OpenCV, cropping/upscaling those regions, deduplicating similar crops, and OCRing the cropped slide images locally. It is a cleaner companion to the full-stage frame deck.

## Reconstructed Slides
![[assets/reconstructed-slides/wsFd22SL1s8/slide-001.jpg]]

- Source frame: `slide-001.jpg`
- Crop: `full` `[0, 0, 960, 540]` score `167.11`
![[assets/reconstructed-slides/wsFd22SL1s8/slide-002.jpg]]

- Source frame: `slide-002.jpg`
- Crop: `full` `[0, 0, 960, 540]` score `167.57`
![[assets/reconstructed-slides/wsFd22SL1s8/slide-003.jpg]]

- Source frame: `slide-003.jpg`
- Crop: `full` `[0, 0, 960, 540]` score `168.42`
![[assets/reconstructed-slides/wsFd22SL1s8/slide-004.jpg]]

- Source frame: `slide-004.jpg`
- Crop: `full` `[0, 0, 960, 540]` score `166.96`
![[assets/reconstructed-slides/wsFd22SL1s8/slide-005.jpg]]

- Source frame: `slide-005.jpg`
- Crop: `full` `[0, 0, 960, 540]` score `172.74`
![[assets/reconstructed-slides/wsFd22SL1s8/slide-006.jpg]]

- Source frame: `slide-006.jpg`
- Crop: `full` `[0, 0, 960, 540]` score `165.23`
![[assets/reconstructed-slides/wsFd22SL1s8/slide-007.jpg]]

- Source frame: `slide-007.jpg`
- Crop: `full` `[0, 0, 960, 540]` score `162.66`
![[assets/reconstructed-slides/wsFd22SL1s8/slide-008.jpg]]

- Source frame: `slide-008.jpg`
- Crop: `full` `[0, 0, 960, 540]` score `163.74`
![[assets/reconstructed-slides/wsFd22SL1s8/slide-009.jpg]]

- Source frame: `slide-009.jpg`
- Crop: `full` `[0, 0, 960, 540]` score `173.98`
![[assets/reconstructed-slides/wsFd22SL1s8/slide-010.jpg]]

- Source frame: `slide-010.jpg`
- Crop: `full` `[0, 0, 960, 540]` score `165.68`
![[assets/reconstructed-slides/wsFd22SL1s8/slide-011.jpg]]

- Source frame: `slide-011.jpg`
- Crop: `full` `[0, 0, 960, 540]` score `165.84`
![[assets/reconstructed-slides/wsFd22SL1s8/slide-012.jpg]]

- Source frame: `slide-012.jpg`
- Crop: `full` `[0, 0, 960, 540]` score `171.69`
![[assets/reconstructed-slides/wsFd22SL1s8/slide-013.jpg]]

- Source frame: `slide-013.jpg`
- Crop: `full` `[0, 0, 960, 540]` score `167.15`
![[assets/reconstructed-slides/wsFd22SL1s8/slide-014.jpg]]

- Source frame: `slide-014.jpg`
- Crop: `full` `[0, 0, 960, 540]` score `167.7`
![[assets/reconstructed-slides/wsFd22SL1s8/slide-015.jpg]]

- Source frame: `slide-015.jpg`
- Crop: `full` `[0, 0, 960, 540]` score `167.13`
![[assets/reconstructed-slides/wsFd22SL1s8/slide-016.jpg]]

- Source frame: `slide-016.jpg`
- Crop: `full` `[0, 0, 960, 540]` score `169.61`
![[assets/reconstructed-slides/wsFd22SL1s8/slide-017.jpg]]

- Source frame: `slide-017.jpg`
- Crop: `full` `[0, 0, 960, 540]` score `169.89`
![[assets/reconstructed-slides/wsFd22SL1s8/slide-018.jpg]]

- Source frame: `slide-018.jpg`
- Crop: `full` `[0, 0, 960, 540]` score `169.89`
![[assets/reconstructed-slides/wsFd22SL1s8/slide-019.jpg]]

- Source frame: `slide-019.jpg`
- Crop: `full` `[0, 0, 960, 540]` score `170.48`
![[assets/reconstructed-slides/wsFd22SL1s8/slide-020.jpg]]

- Source frame: `slide-020.jpg`
- Crop: `full` `[0, 0, 960, 540]` score `169.27`
![[assets/reconstructed-slides/wsFd22SL1s8/slide-021.jpg]]

- Source frame: `slide-021.jpg`
- Crop: `full` `[0, 0, 960, 540]` score `170.26`
![[assets/reconstructed-slides/wsFd22SL1s8/slide-022.jpg]]

- Source frame: `slide-022.jpg`
- Crop: `full` `[0, 0, 960, 540]` score `170.29`
![[assets/reconstructed-slides/wsFd22SL1s8/slide-023.jpg]]

- Source frame: `slide-023.jpg`
- Crop: `full` `[0, 0, 960, 540]` score `166.31`
![[assets/reconstructed-slides/wsFd22SL1s8/slide-024.jpg]]

- Source frame: `slide-024.jpg`
- Crop: `full` `[0, 0, 960, 540]` score `171.19`
## Dense Scene-Detected Slide Candidates
- [[youtube-wsFd22SL1s8-dense-slides]]