---
title: "Slides: 2026: The Year The IDE Died — Steve Yegge & Gene Kim, Authors, Vibe Coding"
category: "slides"
video_id: "7Dtu2bilcFs"
sourceLabels: ["Public YouTube video frames", "Public YouTube metadata"]
---

# Slides: 2026: The Year The IDE Died — Steve Yegge & Gene Kim, Authors, Vibe Coding

## Source Video
[2026: The Year The IDE Died — Steve Yegge & Gene Kim, Authors, Vibe Coding](https://www.youtube.com/watch?v=7Dtu2bilcFs)

## 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/7Dtu2bilcFs/slide-001.jpg]]

OCR text:

> 3
> 
> a

![[assets/slides/7Dtu2bilcFs/slide-002.jpg]]

OCR text:

> °
> Al Developer Trajectory
> t
> 2023 2025
> a code completions (Cop:lot) agente coding assistants (Claude Code, Codex,
> Cline, Amp, Gemini CLI, . J
> | 1 2 3 4
> 2024 2026
> chat based programming (Cursor, Windsurf, Cody! 2
> Siang a We ail know by now that agentic coding is not the Fina Form for Al.
> This talk is about what's coming next: orchestrators.
> et
> ages 2026: THE YEAR THE IDE DIED
> eee STEVE YEGGE = jmp GENE KIM Aprevowstion
> Google DeepMind Software Engineer . Author & Researcher

![[assets/slides/7Dtu2bilcFs/slide-003.jpg]]

OCR text:

> Manual Tools get Automated
> >» NX _ yw
> >» _N Ye
> ; a j a si a
> | are 2 bale Oi
> P aw fy
> ¢ J i 4 -
> 
> Why would we stop now?
> Vans 2026: THE YEAR THE IDE DIED
> Serene STEVE YEGGE GENE KIM
> 
> Google DeepMind Software Engineer Amp . Author & Researcher MPrevowwtion

![[assets/slides/7Dtu2bilcFs/slide-004.jpg]]

OCR text:

> Models are Smart
> i + Enough
> ae 4 t 5 : 7 We are entering a world where Al will be writing and
> . re - de ¥ , newest
> rar eH rf m Ko. 4 reviewing all code.
> e ws eee RE The models don't need to get any better.
> yer! oe . ann:
> a red ‘
> 3 A we Os 2” If the models plateaued, this becomes a 100%
> A F oe i. %
> ‘ , . —
> rs & engineering problem.
> F a Seon . The models are getting better, but for this talk, I'll
> — Wee 4 getting
> assume they won't.
> : 3 . Because they don't need to.
> i po
> [see]
> ages 2026: THE YEAR THE IDE DIED
> ene STEVE YEGGE = jmp GENE KIM APrevournion
> Google DeepMind Software Engineer Author & Researcher

![[assets/slides/7Dtu2bilcFs/slide-005.jpg]]

OCR text:

> 7 e
> We've Seen This Movie el
> Before 7 re _
> Why are so many senior engineers refusing Al assistance? F a
> + watchmakers retused electric tools 4 «> 4
> ea + Caused the Swiss watch industry “Quartz Casis” i nam it
> y) drafters in the 705-805 rejected CAD ane ay
> « photographers refused digital tools oe] z Sie a
> - a ‘ =
> It takes time for people to change. But automation always wins. p> 7 ee
> Viet 2026: THE YEAR THE IDE DIED
> Eee STEVE YEGGE = jmp GENE KIM MPrevo
> Google DeepMind Software Engineer . Author & Researcher

![[assets/slides/7Dtu2bilcFs/slide-006.jpg]]

OCR text:

> 5 Ny The New Quartz Crisis
> a es Me’ cy mm Building software is like Linus Torvalds: One of
> eo ; ao “ \ building mechanical the world's best
> yas ¥@ .——) ’ a@ NV watches “watchmakers”
> /] ow rA\ \ y an old, elegant, handcrafted He says vibe coding isn’t good for
> ff ik < \\ i discipline “seal” work
> | Se, © i oe Same argument from the
> Cm“ “ - Graftsmen, watchmakers,
> 4 , \\ 4-4 iQ = AY = 7 photographers.
> | Na ry nerve
> \ NO ke WT. a Building software the old way will be automated away
> a , am
> 7 and it won't take 10-20 years tha time
> _ net it will take 1-2 years
> AIE/LEAD It Tenex Bloomberg Bre NermULaToss MINIMAX Google
> Google DeepMind larize %<goose ¥', Modal €> turbopuffer Cs %e¢elastic «

![[assets/slides/7Dtu2bilcFs/slide-007.jpg]]
![[assets/slides/7Dtu2bilcFs/slide-008.jpg]]

OCR text:

> J
> cone e
> ’ Adoption Depends on
> —" ‘
> 4 ue = 3” Having UI
> sae ~ “”
> e . wer § « We've established that CLI But coding agents aren't
> eee ets
> : j a = agents are too hard for most heading in this direction.
> nd
> _ devs. Instead, they're doubling
> 7 const st 2 on We clearly need a Ul. down on the CLI.
> Lammers i
> | oo | lu! We also need to lift people up. » How much has Claude Code
> aw om changed since February?
> + Claude Code shoves your nose in
> : onion - everything.
> 54 RES we @ wi 1g.
> jaa ‘ oe” 7 cof ae + You have to watch everything it
> ‘ a zs aoe ae _ does.
> i oe X + We need helpers - models!
> coset _
> Ca fd
> Sea Engineering the future of Al

![[assets/slides/7Dtu2bilcFs/slide-009.jpg]]

OCR text:

> iY .
> : °
> | Coding Agents Should
> 2
> be like Ants
> Z
> 4 One huge problem with coding agents is their
> granularity: You send all tasks, even cheap ones, to the
> expensive model.
> > + "please analyze this codebase for Haws” --> expensive
> + “is my gitignore file still there?” --> also expensive!
> _¢ instead of an ant colony, Anthropic built a huge,
> ~ oa ry el : muscular ant,
> * — There's a better solution: Make small ants.
> 7 ‘
> ce f
> e td
> Sea Engineering the future of Al

![[assets/slides/7Dtu2bilcFs/slide-010.jpg]]

OCR text:

> Coding agents are She deve Vow send thee “unserm ater et ya cote
> Thee (ORAeH ern a Dw orppen lene
> AA eed droge can Get corr SIve Tngy done oN Se mate
> ba Oe rd int wee
> ‘Why are we Lrymg Le Rares 2 vunghe Greer de ever ptheng”
> 
> - - .- a
> 
> Code Summit - o |
> oats cana "
> ie od 3 ely
> Pe ae
> 3 re a
> 7 , f |
> f 3 an bs

![[assets/slides/7Dtu2bilcFs/slide-011.jpg]]

OCR text:

> TheDiverProblem
> ly so big
> tohave astingle diver de everyth
> AlEngineer
> Engineering the future of Al

![[assets/slides/7Dtu2bilcFs/slide-012.jpg]]

OCR text:

> Dealing with Backlash
> “"* Guetcho G. © »
> Mmm right.
> But no.
> My brazenly shameless advise to you:
> Leave the programming to the programmers.
> Do your director stuff instead.
> aml
> eee
> The backlash will be pretty cevere
> We're already seeing it. Just mang you aware now
> Engineering the future of Al

![[assets/slides/7Dtu2bilcFs/slide-013.jpg]]

OCR text:

> IES’ = ical
> , ae
> y ae
> a 7
> |
> |
> |

![[assets/slides/7Dtu2bilcFs/slide-014.jpg]]

OCR text:

> AIE

![[assets/slides/7Dtu2bilcFs/slide-015.jpg]]

OCR text:

> : The Value of Vibe Coding
> as - F: Build the things | want faster
> wo « A: Be more ambitious about things | can build
> A: Be able to build them alone (as opposed to
> requiring other people or a team)
> = F: Have so much more fun doing it
> : O: Have more swings at bat, explore more
> hed : options
> 2026: THE YEAR THE IDE DIED
> eee STEVE YEGGE GENE KIM
> Google DeepMind Software Engineer Amp 1 Author & Researcher Pirevowion

![[assets/slides/7Dtu2bilcFs/slide-016.jpg]]

OCR text:

> Booking.com (2025) ~~
> | * Bruno Passos, x
> a Group Product a
> Manager, Developer Bases
> Experience ——E *
> - Laura Tacho, CTO, oo
> DX wed . Daily Al users ship 50%: more than non-Al users
> c$25 1 — ..
> fm | : nl
> , Source blips sated srevoknon com watch’ 1121936057 oes 4.
> 2026: THE YEAR THE IDE DIED
> eer STEVE YEGGE GENE KIM
> Google DeepMind Software Engineer >} Amp . Author & Researcher MPrevourtion

![[assets/slides/7Dtu2bilcFs/slide-017.jpg]]

OCR text:

> wee ~CiSCO (2024) = | ,
> , ry UB Ea o~
> _ ; - John Rauser, Senior fin BN Be EG
> / Director of Software 7 _—~
> Engineering as |
> - Anand Raghavan, a a
> Senior Director of ‘Ea. ———
> Engineering, Al 1 2 , : ag
> 6 Ld
> a "
> an Source hilps shades, ra vohihon com watch’ 1007959632 | Hef. smoroemen ees erence art —
> 2026: THE YEAR THE IDE DIED
> ene STEVE YEGGE GENE KIM
> Google DeepMind Software Engineer 3} Amp . Author & Researcher MPrevowmion

![[assets/slides/7Dtu2bilcFs/slide-018.jpg]]

OCR text:

> aa ~Vibe Coding Workshop For Leaders
> CZ
> ioe
> ees oo oe” f
> Engineering the future of Al

## 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.