Slides: No Vibes Allowed: Solving Hard Problems in Complex Codebases – Dex Horthy, HumanLayer
Source Video
No Vibes Allowed: Solving Hard Problems in Complex Codebases – Dex Horthy, HumanLayer
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
OCR text:
ae
| |
ae: \ 3
a a tS
OCR text:
Yegor’s Talk:
=e 100k Devs across all company sizes
Most AI in Software Eng leads to
rework
me Re ee ee cee adexherthy coc td
NO VIBES ALLOWED: SOLVING HARD PROBLE
en eee DEX HORTHY /Foundera ceo HumanLayer
OCR text:
—— . Software Engineering Productivity Increases from Al Use ,
Orientative Guidelines
c +10-15% +0-10%
2 Complex tasks require deeper Constrained by outdated code &
2B human insight intricate dependencies
aL x
@
a
E .
°
0
¥
Vey +35-40% +15-20%
3S These tasks are often repetitive Legacy projects still benefit on
and well-defined simpler tasks
—P Greenfield Brownfield
Project Maturity
AIE /CODE ODEBASES NO VIBES ALLOWED: SOLVING |
en eee DEX HORTHY /Founcera.ceo ~HumanLayer
OCR text:
reallyf..
Never going back
hard
ovibes allowed|adrthy
Code Summit
AIE
OCR text:
Slightly smarter - intentional compaction
[__“Syeten taotreetione sym ratrvetionn 1
a C ae : aati =
( —aeneahe ie |
( wis} | Intentional Compaction *— ois)
[Wear enaneye —_i=—=~SC~*«~<
a "Write everything we did ( Rens wreyenned
ye C7 so far to progress.md, ( Rewdl? }
a j ensure to note the approach oc
— | 1 Asabatant messane we're taking, the steps | progress oe )
e ae we've done so far, K LD
—r and the current failure ( Save) )
oS | we're working on” Cai?)
Ca (a)
Terie weed 6c (___—~*d
a. rn a worsens t
NMRA prearase te pregrane ot 0 X J
| . us . 4
Aepee
no vibes allowed | a@dexhorthy Code Summit |
athens Tamer Mio] acm |-iel [unl (tama Xen Kel:) AMV KG01 1-10] Mm 00 Kee egy
Google DeepMind 88 Cognition & FACTORY @Graphite sneo4j Yauth@ > worko
OCR text:
owed
adexhorthy
Code Summit
AIE
Code Surmit
Googeaodied
OCR text:
a
- a ae | a: i a
| | =) |
_ tha; ems a
OCR text:
“3 ai that works: Advanced Context Engineering for Coding Agents
A deep dive into managing research, specs, and planning to get the most out of coding agents and coding CLis
Video (1h27m)
i uO Hyper Engineering
| - fel iaF [=
we
&
® ai that works #17 [a] ic r
es)
NY Pd beewe
no vibes allowed | adexhorthy Code Summit |
Engineering the future of Al
OCR text:
Challenges With this approach
. Either too long or under-specified
| It gets out of date
—— so vibes allowed a&dexhorthy foc tad
Engineering the future of Al
OCR text:
9 Mitchell Hashimoto 's)
/ - My favorite part about is that you can share your whole
session globally. PRs with Amp threads attached make me very, very
happy as a maintainer. here is one from a bug fix this morning:
ESA
as ee Ps ape ¢
a no vibes allowed &dexhorthy see EST
Engineering the future of Al
OCR text:
wed | adexhorthy mn
Code Summa
wm | | fi
aan —
—— ———
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.