---
title: "Slides: How Lovable self-improves every hour — Benjamin Verbeek, Lovable"
category: "slides"
video_id: "KA5kPbdkK2E"
sourceLabels: ["Public YouTube video frames", "Public YouTube metadata"]
---

# Slides: How Lovable self-improves every hour — Benjamin Verbeek, Lovable

## Source Video
[How Lovable self-improves every hour — Benjamin Verbeek, Lovable](https://www.youtube.com/watch?v=KA5kPbdkK2E)

## Relationship To World's Fair 2026
These slides are extracted from a public AI Engineer YouTube video connected to World's Fair 2026. Speaker-matched clips are supporting context unless later confirmed as exact session recordings; official livestream recordings are day-level/event-level source material.

## Related Scheduled Sessions
- No individual scheduled session mapping has been assigned yet; treat this as an event livestream deck.

## Extracted Slides
![[assets/slides/KA5kPbdkK2E/slide-002.jpg]]

- Recreated text/layout view: [open HTML recreation](/assets/slide-recreations/slides/KA5kPbdkK2E/slide-002.html)
- AI slide classifier: `title_card` confidence `0.98`
- Text source: agent_vision.

Slide text:

> How Lovable Self-Improves Every Hour
> Benjamin Verbeek
> Lovable

![[assets/slides/KA5kPbdkK2E/slide-003.jpg]]

- Recreated text/layout view: [open HTML recreation](/assets/slide-recreations/slides/KA5kPbdkK2E/slide-003.html)
- AI slide classifier: `title_card` confidence `0.98`
- Text source: agent_vision.

Slide text:

> The Holy Grail:
> Continuous learning at scale

![[assets/slides/KA5kPbdkK2E/slide-004.jpg]]

- Recreated text/layout view: [open HTML recreation](/assets/slide-recreations/slides/KA5kPbdkK2E/slide-004.html)
- AI slide classifier: `content_slide` confidence `0.99`
- Text source: agent_vision.

Slide text:

> DESIGN & DEVELOPMENT
> Now is the time to build something lovable

![[assets/slides/KA5kPbdkK2E/slide-005.jpg]]

- Recreated text/layout view: [open HTML recreation](/assets/slide-recreations/slides/KA5kPbdkK2E/slide-005.html)
- AI slide classifier: `content_slide` confidence `0.98`
- Text source: agent_vision.
- OCR decision: ready — Dense embedded screenshot text is better handled by OCR.

Slide text:

> Engineering the future of AI

![[assets/slides/KA5kPbdkK2E/slide-006.jpg]]

- Recreated text/layout view: [open HTML recreation](/assets/slide-recreations/slides/KA5kPbdkK2E/slide-006.html)
- AI slide classifier: `content_slide` confidence `0.97`
- Text source: advanced OCR `rapidocr-live/border-trim/contrast`.
- OCR decision: ready — Dense embedded UI screenshot with small text.

Slide text:

> Lovable Main Agent [venting...]APp 23 Mar at 4:42 PM
> Agent Vent
> AIE code--copy consistently fails for user-uploaded fles with spaces in the filename (e.g. "Screenshot_2026-03-23_at_9.52.56 AM-2.png"). Tried both raw spaces and URL-encoded %20. Only fles without spaces in their names (e.g. "PNG_image-2.png") copy successfully. This blocks
> using user-uploaded screenshots in projects. The files ARE visible via lov-view (renders the image),
> but copy always says "source fle does not exist".
> Almessage| Braintrust|PostHog Project: 985c0363-d79e-42fa-b548-38c5213503c5
> Ategree AlEngineer
> EUROPE

![[assets/slides/KA5kPbdkK2E/slide-007.jpg]]

- Recreated text/layout view: [open HTML recreation](/assets/slide-recreations/slides/KA5kPbdkK2E/slide-007.html)
- AI slide classifier: `content_slide` confidence `0.98`
- Text source: advanced OCR `rapidocr-live/full`.
- OCR decision: ready — Embedded code-review UI screenshot with dense text; OCR is better for the small content.

Slide text:

> AIE Nicis-Degrandet swarmia APp 2 Apr at 9:15 AM
> 3fileschanged1690.Merged
> BenjaminVerbeek approved
> 8comments by Niels Degrande,Benjamin Verbeek,and 1 other
> bLovable
> Engineering the future of Al


### Hidden Non-Slide Evidence
- [`slide-001.jpg`](/assets/slides/KA5kPbdkK2E/slide-001.jpg) — `speaker_stage` confidence `0.99`; Stage photo with speaker and audience; not a readable presentation slide.

Classification audit: `raw/sources/slide-ai-classification/slides/KA5kPbdkK2E/audit.json`

## Slide-Derived Subjects To Review
Subject extraction uses video title, related session titles/descriptions, transcript context, and OCR text when available. OCR is best-effort and should be reviewed against the embedded slide images.