---
title: "Reconstructed Slides: Recsys Keynote: Improving Recommendation Systems & Search in the Age of LLMs - Eugene Yan, Amazon"
category: "slides"
video_id: "2vlCqD6igVA"
sourceLabels: ["Cropped public YouTube video frames", "Local OpenCV slide-region detection", "Local RapidOCR"]
---

# Reconstructed Slides: Recsys Keynote: Improving Recommendation Systems & Search in the Age of LLMs - Eugene Yan, Amazon

## Source Video
[Recsys Keynote: Improving Recommendation Systems & Search in the Age of LLMs - Eugene Yan, Amazon](https://www.youtube.com/watch?v=2vlCqD6igVA)

## 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/2vlCqD6igVA/slide-003.jpg]]

- Recreated text/layout view: [open HTML recreation](/assets/slide-recreations/reconstructed/2vlCqD6igVA/slide-003.html)
- AI slide classifier: `content_slide` confidence `0.98`
- Text source: none.
- OCR decision: ready — dense diagram and small text across multiple callouts/screenshots
![[assets/reconstructed-slides/2vlCqD6igVA/slide-004.jpg]]

- Recreated text/layout view: [open HTML recreation](/assets/slide-recreations/reconstructed/2vlCqD6igVA/slide-004.html)
- AI slide classifier: `content_slide` confidence `0.93`
- Text source: none.
- OCR decision: ready — slide contains a dense screenshot plus a readable aside title

### Hidden Non-Slide Evidence
- [`slide-001.jpg`](/assets/reconstructed-slides/2vlCqD6igVA/slide-001.jpg) — `sponsor_logo` confidence `0.99`; sponsor logo wall
- [`slide-002.jpg`](/assets/reconstructed-slides/2vlCqD6igVA/slide-002.jpg) — `title_card` confidence `0.97`; title card

Classification audit: `raw/sources/slide-ai-classification/reconstructed/2vlCqD6igVA/audit.json`