---
title: "Slides: Real-time Experiments with an AI Co-Scientist - Stefania Druga, fmr. Google Deepmind"
category: "slides"
video_id: "wNH3q9pqn0U"
sourceLabels: ["Public YouTube video frames", "Public YouTube metadata"]
---

# Slides: Real-time Experiments with an AI Co-Scientist - Stefania Druga, fmr. Google Deepmind

## Source Video
[Real-time Experiments with an AI Co-Scientist - Stefania Druga, fmr. Google Deepmind](https://www.youtube.com/watch?v=wNH3q9pqn0U)

## 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/wNH3q9pqn0U/slide-001.jpg]]

OCR text:

> aWS
> 
> eee)
> @®Graphite WW Windsurf 4 MoneobB
> Mdaily £3 augment code WorkOS

![[assets/slides/wNH3q9pqn0U/slide-002.jpg]]

OCR text:

> — _
> F
> Re Ee
> re. = | 2. o—
> s j a a
> e
> W@ O°. 8 GEBSE “OO #
> Lament
> | a

![[assets/slides/wNH3q9pqn0U/slide-003.jpg]]

OCR text:

> -: “Semen te ores 6 ce me,
> | 7 = Labeend x... ds
> SPT COMM Tl OBB t ween .
> ¥
> : Ta
> Se
> i & Seneineits x
> aD
> - _ - Q
> _ et 38-0° 86 Gease"oo 8
> 1a
> _—— re WAYS
> ed

![[assets/slides/wNH3q9pqn0U/slide-004.jpg]]

OCR text:

> = §=~S
> = ; -
> a
> aE ;
> ; 1
> 7: Olga smh FT /iw]--Bela)? ae |
> . r. a Microsoft @yr{?
> ! 4 | at aN ai '

![[assets/slides/wNH3q9pqn0U/slide-005.jpg]]

OCR text:

> : ; on Re re gp mene oe te ones ve oe
> | eT BEE * r | fr 2 ewe A eae
> ¥ »—— ee OT 2 trans Anantare
> eee. a mag ryan oni econ
> * | | | } _
> Se
> ' ———t oa ° ran - lees
> we eet zz r a
> , 4
> — r I 3
> 38-0" 6 GOuasee*oo 8
> aWws
> a 7 Come |

![[assets/slides/wNH3q9pqn0U/slide-006.jpg]]

OCR text:

> Real-time Al Co-Scientist
> Input Sources Backend Services
> bc
> Jacdac USB LabMind App
> (OD + O sere stoaue | aX
> comers (5) wean rt
> Q > Drewmn “a -
> ame y,
> a eens an ey os
> AB
> A yr
> ies aws
> 7 , Far t ee a =
> > ~~

![[assets/slides/wNH3q9pqn0U/slide-007.jpg]]

OCR text:

> Ingredients
> ses
> i ; c a 4 Sensor Array
> = : :
> seesees is ‘ *c humidity
> s* @ ry )
> j a ; ~ -
> A : % Processing
> Uh 2
> q Z Response
> weouss APE
> omen Microbit boards, Jacdac Sensors, Recamera
> air aie a Microsoft ary?
> ; a. ;

![[assets/slides/wNH3q9pqn0U/slide-008.jpg]]

OCR text:

> Fast and Safe Experiments
> n= , an
> wk. |
> as a aes
> “— ) - -, pond 7 ea
> pe er
> oul a Microsoft §=ooou®
> . nm, a Microso
> mee}

![[assets/slides/wNH3q9pqn0U/slide-009.jpg]]

OCR text:

> AIE
> Lab in a box
> (kind of)
> aws

![[assets/slides/wNH3q9pqn0U/slide-010.jpg]]

OCR text:

> SOLUTION NUCLEATION GROWTH FINAL CRYSTAL
> 
> Demo 1: Crystal Growth = — #
> The Science of Supersaturation |)" @)7 B*&
> When salt dissolves in hot water, we create a
> supersaturated solution - more satt than water can
> normalty hold at room temperature. As the solution
> cools and water evaporates. salt molecules have
> nowhere to go axcept to arrange themsalves into ee.
> organized crystal structures. The seed crystal provxies
> a template, and layers build up atom by atom. 4 21 3294
> Key Scientific Concepts:
> 
> - Solubility curves: More salt dissolves in hot
> 
> water than cold
> - Nucleation sites: Where crystals begin to form bo
> - Growth rates: Affected by temperature, a
> concentration, and purity
> Se Pe
> D pattern
> “a
> a oe _ aws
> ! yf : ed
> a 1%

![[assets/slides/wNH3q9pqn0U/slide-011.jpg]]

OCR text:

> Demo 1: Crystal Growth
> bursts
> : ; * Critical supersaturation reached — sudden
> e 4 o- crystal seed formation
> 124-52 C-) + Energy barrier overcome ~ rapid cluster
> formation
> 1 ye Cryad termatinn Detection (Light le attereng)
> ‘ , ‘ aesprss te rule re °
> —
> Io
> zie ae A
> eae w Microsoft QO)
> , a Un
> LY ‘N
> on ‘+.

![[assets/slides/wNH3q9pqn0U/slide-012.jpg]]

OCR text:

> .
> Demo 2: Yeast Fermentation
> Yeast are living microorganisms that eat sugar and —-
> produce CO, gas as waste. In bread dough, this CO, @ Webcam & @ 6 B @ Wedcam s oBa
> gets trapped by gluten proteins, creating bubbles that ~—
> make the dough rise. This is anaerobic fermentation -
> the same process used to make bread, beer, and wine a
> for thousands of years.
> Key Scientific Concepts: A y |
> + Enzyme activity: Yeast enzymes break down
> complex sugars a oe
> « Metabolic rate: How quickly organisms process 4 AN
> nutrients a ne e
> + Gas tawe: CO, production creates pressure and
> volume changes
> * Growth phases: Lag —- Exponential -* Stationary -+
> Decline
> \ 5
> . > ae a7
> 0 A
> \~ >?

![[assets/slides/wNH3q9pqn0U/slide-013.jpg]]

OCR text:

> Demo2:Fermentation
> AIE
> Activity vsTe
> Enzyme Actvity
> Ag:27.6℃
> AchualRange
> Fermentafion experiment recording speedup
> Microsoft
> smol?

![[assets/slides/wNH3q9pqn0U/slide-014.jpg]]

OCR text:

> Open-Source Ecosystem
> [| Shieid_| _—— — . . ; 7 é 2. ea me Pw)
> Tl ‘ ‘oa ba ri i a % e
> ~ 18 aa = Pi Ps i
> > Wii Ih ee ad
> emia = i Rien . - : Pathways to Open-Source
> a Peston aa
> Open Trickier bape.” UEDoyLempte photoreactor Jvtiiee moten plaitorm
> gtind omrenmers! pape Mubteesdca eve
> ndeaietion . -] R
> Vi” | li ,
> 4 ; 5 i oe
> OPEN SCIENCE
> Mb isis Open Bioreactor oe
> ¥ PANTHEON Com wore
> b< 4
> bee 7 a ¥
> a, a Microsoft ary?
> on _<e |

![[assets/slides/wNH3q9pqn0U/slide-015.jpg]]

OCR text:

> Al Engineers and Scientists Coming Together
> 7 ar gar of Droplet Magic!
> An is Gig Ee y ¥
> Lina = a> 5
> | in : i. ie - *
> aa “= —  %& = Sg . ee
> Bate ye eR
> an A _ ae | if : ~— i
> 2y 1 ena \ i 3 : a <a F
> / . ‘.
> P Bemucage PEST), Keith Brown @osten University), Josh Smith LW), Jered Nekahare (AW)
> Pathways to Open-Source Liquid handling and mixing with robot arm
> Hardware for Laboratory ij *
> Automation ia :
> ty zg. , ;
> a a
> ts.washington.edu/machnes/scienceautomation/ = my ~— No :
> Y
> = 7 aire —_
> . t | s

![[assets/slides/wNH3q9pqn0U/slide-016.jpg]]

OCR text:

> The Future... ow vee
> f
> @ cols . Je ee
> _ WR a 98
> ©) Soir Cima -oO8 oe — meeo oa
> AG 7 - . . / -_ .
> a
> : ae — +}
> iit
> ae aws
> i t fe oan ed

![[assets/slides/wNH3q9pqn0U/slide-017.jpg]]

OCR text:

> wh
> an
> 4 y
> | ; 7 J | +e le
> . ™ | ‘

## 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.