Slides: Ship it! Building Production Ready Agents — Mike Chambers, AWS
Source Video
Ship it! Building Production Ready Agents — Mike Chambers, AWS
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:
aWS
ee)
@®Graphite W Windsurf MongobB
Mdaily £3 augmentcode WorkOS
OCR text:
Shi he at
Y aan
Building Production-Ready Agents a A sd f
3 Sed 4
an
oS " 7 :
ae ee a
9 a “ae ’
_ ‘4° . am
io or ee
ae a
oe w a
Mike Chambers (he/him) wo —
:
Teale a Ts Advocate, Al Engineering @ AWS e , aws
al
i Vi fai)
, a Microsoft § Smo
OCR text:
Learn the fundamentals
of generative Al for real-
world applications | 5
eRe cect a ee eA te eno) ET
ncn eccpel enna rere 4 bY -
penarion qa
Ld
ohn.
eS
dl
|
; Eva
= Vr 9
a a Microsoft § Cl)
OCR text:
at
WW eee
\ Wt. 2 > hh |
Sy lO
| ® ° Sg Daas )
| | a Microsoft Ssmmoln
OCR text:
a ae Eh ee cS a tat sso ac | ea aad a aS
oe < ate-wf-2025-demos res de . = 7
oO ae od i nn rrr
fener a) Grae ar ao em Cea
se mi ny
2 ; 2g
AS
m4
ors
roll_dice(sides: - od a
mS |
ag vi | 5 rath Otol ae eeee eens
3
ers 9 : 7 i :
os rnit (seit, models a is
so self madel - cadet
si seif.taois = ¢ : roli_dice}
oa) | Tone spm relat o1 ac Tr Ret ot ars DS eel a aces
Q is
io ee get system ororptisett) - > cris
parse tool cavitsulf, cespenser ote) => aoa - i | ao
, a nt 3: master? = G1 0.%0 0 ts) AWS: profile:m:kegc-Admin par Naar Pcelane) nats 71 0
OCR text:
ad rd ate-wf-2025-demos re aS Ty. a a
1m Pyke Om nn sone
Foxerr rama C1 Grand Mar oy, Seale eraeraaas get system Prompt
LP 9 : bot te _
15 | get_syster_prorptiselt) => ccd ee
fe we ||
Pry avi
ae)
tS
cs 20
rae
oes 22
Pa]
ray
25
mm
27
7 28
; a nt i- master? “=: 05 0.°,0 i) AWS: Piece > ‘Amazon Q a Rare: rata vwenv) oo.
OCR text:
°° re PT ROL EP Ata SrolTa ster re nn eran §
i mtn x ran ane ee
Foren tmraiel aes mar oy Sraa racers get system prompt
© i) ee ee
15 (ek en ee amie aoe by ce ee
28 _
oe |
4 :
38
S cs
ga coe | parse_tool_calilself, response: ae Pace » | ao
41
uv PROBLEMS OUTPUT TERMINAL — CODE REFERENCE LOG IE AERTS cc
PU ete [ae eer Vstee (on Mee ew A Cle Deer te are
mikegc local-agent *% uv run main.py
# Simple Ollama Agent Demo
Q ATs mane SMR CORT OOS
Try: ‘roll a d20', ‘roll a dice’, ‘roll a di2' |
dL for initiative and add a dex mod of fj
wm DSP ie tt eee Oe © Ee MOREE CCS sleoli] uit ye Come o(aitia} paras Paelane) 3.12.8 {\venv': venv) 7
; a Simol
+
OCR text:
Agent
, Mode! |
| a :
7 aws
|
OCR text:
AIE
LLM
Microsoft
smol?
WordsFair
OCR text:
3 Amazon Bedrock Agent
Instruction Pvoib ora co18] 0)
r “You are...” IN
és PSUR neon]
aes)
Acton Group
ENT RSB EE Taek)
Rd ty Pp
ir . a
laa w Microso
OCR text:
Cae 1c) ne 9 Go EO CT
6 Amazon Bedrock 9
Amazon Bedrock < Overview ut
% Getting started .
cervew Model spotlight
Provaters A\
¥ Foundation models
Model catalog Ni a
statog New Anthropic's Claude
Marketplace deployments: New
Choate the exact
Custom models (Datilavon, Fine POPC MTCC combination of
tuning Continved pee training) IMaiigence, speed, ard
Imported models & bean “ ae rae ee cost to wet your nceds All
of the latest Claude
Prompt Routers models, ithe Clade 4, ace
Rar are PR SIE
ow founds ie aa) edal umiaa . available in Amazon
Playges re eee ne Bedrock
Chat / Text
Open ia chat playground
image / Video
Builder tools
Agents
Flows ~‘ <>
Koomtetor Bases
con fc
onan Jf
OCR text:
Create agent x
Name
TOT TTC
\ game_master agent i
Waal. Maa teve ea ATO brates Fong grand Preasmer tae gen aT
sates
Description - optional
Tre tee Sane ST naracens
Multi-agent collaboration
beara more about Tul agent colatoraton (2
* Enable multi-agent collaboration.
Bc Mage HR ee Bye TT aT a a nme Meer ae ote he
daca
cn Ca
en “
OCR text:
Select model x
QQ Serch ers late mantels aed erence :
Bedrock Agents optimized Learn more (2
1. Categories 2. Models 3. inference
Model providers ‘ Models with access (4)
_ & amazon Nova Breniier Select model to show
* anference options.
A\ Anthropk
+ Neva Pro
Nova Lite
Nova Micro
QO Cort tnd the mote, you are icobkurg (or? See al modets here (2 Cancet Sash
be *,
“se. ua
OCR text:
wn HP. ico ee 9 Go ET OC
e Amare Redrock > Agents ) game master_ageat > Agent builder: game_master_agent @ oO 6
© reste and use a new service role
Amazon Bedrock < C} Use an easting senace role Test Agent Od.” >?
¥ Getting started Useng OOT Chan
Overwew Select modet
Provaters
A\ Claude 3.5 Hailkue? OF
Foundation models On-demand
Model catalog New
Marketplace depleyments New BL PURSE SSC PURPURA EE
Custom models (Dstalation, Fine PA pe Cee Wye tet trae
tining.Centinued pre‘eraining) You are a games master who can help me play ttrpg games
imported mode's I
Prompt Routers,
* Playgrounds
Chat f Text & Additional settings
image / Video
é nevi
© Buitder tools . ‘pter your messoge here
Agee Action groups (0) ir bate
flows l Q batectoe gongs . :
Kooptedoe Bases ; ©
( Run
Name 7 Rescrintinn State 2 tastontited 9
Be aa
“e i ° oO
ite g Microsoft
OCR text:
Mike’s prompt template... a
re 4
f tore OLE
AE “scoot 0: "Ask Mike about {topic}", lyestT
“Gok ees [
“cloud scale MCP servers",
"open source SDKs for developing model-first agents",
“where I can get an IRL d20”,
Nan aS
] aa
: Ree
ho
bos
an
core Be
AAS cai)
a a Microsoft =o
OCR text:
«6S rea . te
ae a» i
| World's Fair |
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.