Slides: Building Cursor Composer – Lee Robinson, Cursor
Source Video
Building Cursor Composer – Lee Robinson, Cursor
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:
|
|
_ |
OCR text:
Composer combines coding intelligence with
: best-in-class speed
; Best Open
Fast Frontier
Frontier 7/2025
Composer I
Best Frontier
0 20 40 60 0 100 200
_ intelligence Speed
, 4 q GB composer (Cursor Bench score) (Tokens per Second)
wh N
BUILDING A FAST FRONTIER MODEL WITH RL
ee LEE ROBINSON / vp, Developer Education QJ CURSOR
OCR text:
: 3%
oa bd
ate,
Ste,
hy a, e
Pe eT N
tae - ;
re is wt 7
Spey “ae : ;
#6 Ee
ae ry
. ; ‘
‘ty, . YY
Sean, Ph tg hee . | r
i a
, o
a
| 5
-t i
OCR text:
ee
ae:\:
"<< — :' ha
OCR text:
; Agent RL @ dskspeceEtrotFunc should be calied mth
®e
aw -@
Roll 1 Read Edit
, nowt File File
Rollout 2 “ee Read Edit
; Seu File File
-
Google DeepMind vy auth® €} WorkOS x-Sourcegraph ¢ stigg Weights & Biases &
OCR text:
7
_ te
OCR text:
M POSEr Uses more Search ang ire Poon Li]
file Fed tools lative to code (i
Updates. Sern
~ i
are : aed
OCR text:
y a Se . ee
co 1 - <q,
_ a
BUILDING A FAST FRONTIER MODEL WITH RL
en ee LEE ROBINSON | vp. Developer Education QJ CURSOR
OCR text:
will brown
Anthony
Bwillccbb
ok composer-1 is pretty nuts, and the code it writes is quite nice.
probably my new dailly driver formany things
sync instead of async,and very quickly iterate on fixes.follows
instructionsverywell
jessald
piessafo
Composer1in Cursoris sofast.We arenowin theride the tigerphase of
softwaredevelopment.
AIEngineer
Engineering the future of Al
OCR text:
ra
a
OCR text:
|
|
Ste —
id
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.