Slides: Agents vs Workflows: Why Not Both? — Sam Bhagwat, Mastra.ai

Source Video

Agents vs Workflows: Why Not Both? — Sam Bhagwat, Mastra.ai

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

Extracted Slides

slide-001.jpg

OCR text:

aWSs

ee)

@®Graphite W Windsurf MongobB

Mdaily £3 augment code WorkOS

slide-002.jpg

OCR text:

Wholam Why I'm giving this talk

~-f.*

\ ava ) Mast

Sam Bhagwat 4 The Typesc :

Tee aerten CET Str Fey] bin

a npm create mastra

|

slide-003.jpg

OCR text:

In December Anthropic

wrote a great blog post

that canonically defined

agents and workflows

- | a Microsoft @yr{??

slide-004.jpg

OCR text:

In April OpenAl A practical

released a paper building agents

on the topic

hy:

a a Microsoft Qa?

slide-005.jpg

OCR text:

@openai

AIE

Hot Take #1:

HELLO

my nameis

ThatGuy

Don'tbe That Guy

github.com/mastra-ai/mastra

Microsoft

smol?

World'sFair

slide-006.jpg

OCR text:

Sometimes That Guy works for a

FAANG type company in a public

facing role

Then the rest of us are really in for it

_ aws

| ara) | name

slide-007.jpg

OCR text:

“Use the platform” was a codeword for

why React was wrong and anti-web

4 aws

== nana

slide-008.jpg

OCR text:

Some of our users loved this.

But many folks didn't

x | a Microsoft @yry{?

slide-009.jpg

OCR text:

So when | see APIs :

like this, it gives me

(bad) flashbacks

a | a Microsoft @yr{?

slide-010.jpg

OCR text:

AIE

Or like this:

const [notifyTeam,assignPlan]= await Promise.all（[

step.run("notify_team",async（)=>{

return{notified:true};

step.run("assign_default_plan",async()=>{

return{planId:“starter"};

github.com/mastra-ai/mastra

Microsoft

smol?

slide-011.jpg

OCR text:

AIE

Or like this:

const [notifyTeam,assignPlan]= await Promise.all（[

step.run("notify_team",async（）=>{

return{notified:true };

step.run("assign_default_plan",async()=>{

return {planId:“starter"};

github.com/mastra-ai/mastra

slide-012.jpg

OCR text:

A Pattern Language. aE eae

‘Towns -Bedings ‘Constroction L: as , et =

ee } is

a x ar G ae . .

Ay .

<a

0 erm — a -

Sara Ishikawa - Murray Silverstein - al

Miax Jacobson - Ingrid Fiksdahl-Kiag

aws

~~] ~ |

slide-013.jpg

OCR text:

a . |

How the WorkOS folks put it in their a ar

Mastra workshop yesterday:

Popents are stateaat A ent ten teat Dar LAs

ce renner

e Make cies sous hasedion context e Wa cdate mpouts Gutpats with Zo

esi tis te mn ete Sasas Cs ie Mey clever)

i Ot ea Cua oe CO Cee Ce Tr, cea 18 AD Gaara

irae rt LAL ed a Pe SOs ro en ae me Ce eee)

an a Microsoft = (aU°

slide-014.jpg

OCR text:

Here’s another way to put It:

(1) Agents are a turn

based game.

(2) Workflows are a rules

engine for your tech

Ugeto1

; sick ES 10 Uh

slide-015.jpg

OCR text:

Here’s another way to put It:

(1) Agents store threads of messages and

continuously interact with users

(2) Workflows can have branching. parallelism,

conditions. loops, can suspend/resume, etc

a, aws

! 7]

ee

slide-016.jpg

OCR text:

At the end of

the day it’s all

just a tradeoff

Peer err eae

ar | -

i a Microsoft ary?

slide-017.jpg

OCR text:

Rules of agents + workflow composition

(1) Agents have tools

(2) Workflows have steps

(3) An agent can be a step

(4) A workflow can be a tool

(5) An agent can be a tool

(A\ A workflow can be a step

i aws

ee = | See

slide-018.jpg

OCR text:

const researchAgent nev Agent（{

nane:'research-agent',

Lnstrocttons:You are a research agent that analyzes

Agent supervisor

questtonsopenai(gpt-4'),

const summaryAgent-new Agent（{

nase:'sunary-agent',

instructlons:You are a su

Supervisor

model:openal(gpt-4).

AIE

const researchTool=createTool（{

id:'research',

execute:async（tnputData））{

constresult=await researchAgent.0

rate(inputData.query);

return{text:resutt.text）;

F.

const summaryTool=createTool（{

ld:*summarize'

execute:async（{inputData}）{

const result -awalt summaryAgent.generateLnputData.text);

return{text:Tesult.text）;

export const supervlsorAgent=new Agent（

hane:'supervisor-agent'.

nodel:openal('gpt-4'),

research and summarization tasks.

tools:[researchTool,summaryTool],

aws

Word'sFair

slide-019.jpg

OCR text:

Workflowastool

export const agent = new Agent（{

name:'Agent',

instructions:Ask the user for

AIE

their location,check weather，

and plan a trip，

model:openai('gpt-4o-mini'),

workflows:

checkweather,

planTrip

memory,

github.com/mastra-ai/mastra

Microsoft

smol?

World'sFair

slide-020.jpg

OCR text:

mastra

AIEnginee

World'sFair

slide-021.jpg

OCR text:

AlEngineer

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.