---
title: "Slides: Building AI Agents that actually automate Knowledge Work - Jerry Liu, LlamaIndex"
category: "slides"
video_id: "jVGCulhBRZI"
sourceLabels: ["Public YouTube video frames", "Public YouTube metadata"]
---

# Slides: Building AI Agents that actually automate Knowledge Work - Jerry Liu, LlamaIndex

## Source Video
[Building AI Agents that actually automate Knowledge Work - Jerry Liu, LlamaIndex](https://www.youtube.com/watch?v=jVGCulhBRZI)

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

OCR text:

> INNOVATIONPARTNER
> aws
> PLATINUMSPONSORS
> Graphite
> WWindsurf
> MongoDB
> daily
> augment code
> Workos

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

OCR text:

> Al Agents can “Automate Knowledge Work"
> A big promise of Al agents is making a me eg ey xe ‘
> knowledge workers more efficient:
> iweancuw yom a =
> e Lower cost, save time S23 oes
> e More data, better decision-making — Aa a Pave
> on your processes
> Every b2b SaaS is talking about it. ——EEE—ee
> | SFO Ee PaO oe od
> But does “knowledge work automation” just -
> mean RAG chatbots? - 7 @
> Le Work Alforol sir
> ap --
> ieee 2 .
> 2
> 7]

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

OCR text:

> The Alpha is in Unstructured Data
> 90% of Enterprise Data Lives in
> Documents*
> Humans historically needed to read/write ‘i "
> juman
> these documents.
> For the first time, Alagents can reason ets \ \
> and act over massive amounts of . —s a Ry
> unstructured context tokens. Agent
> 3
> i a al)
> a Microsoft §=SUOOU

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

OCR text:

> Special Release: Excel!
> We built an Excel agent capable of:
> 
> 1. Data Transformation: Transforming wore mmene mrs mmm = . . a
> each sheet into a normalized 2D CEN. 0 Teaiela oua pares: he
> format. —_
> 
> 2. Agentic QA: Answer questions over binds Minit L.
> cells in any of the sheets. et 5
> Sen to
> ieee] 0 Ee
> This is available gy TS a ris, ce :
> in early preview. sy fu 4. ® . as
> Ae CoD
> ol Ld 2 SG! ame ome awe eee
> ld “ty. she | 5

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

OCR text:

> Automation UX
> Use Cases Be aie =
> @ Financial Data Normatization ee ee testers nee
> @ Datasheet Extraction °
> @ Patient Record Extraction .
> @ invoice Reconciliation ee
> bttps./grthub com‘run-lisma invosce reconciber
> mere T, rai)
> a Microsoft =O

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

OCR text:

> AIE
> Document
> Agent Use Cases
> Real-world use cases
> of workflow automation
> aws

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

OCR text:

> [Automation + Assistant UX] Financial Due Diligence
> CARLYLE
> Use Case
> An e2e leveraged buyout agent
> Impact
> “This end-to-end agentic
> workflow to do a leveraged
> buyout model created
> decision-making value in the
> tens of millions of dollars.”
> LlamaCloud
> Excel Normalization
> Agent
> Document Extraction
> Agent
> SQL Tool
> Vector Retrieval
> Tool
> File Tool
> LlamaIndex
> Ask your Excel agent
> Due Diligence Copilot
> 28

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

OCR text:

> 4
> , aa
> The Leading Platform ClO [OR rosa (al -ms x0] 000] a[ cit 0]0 RU: Sumn ameEs EES EEE?
> for Document Al
> Llamalndex :5 tne most accurate and Custamizable “~—
> ee ee Ce ae.
> with agentic Al,
> Backed By: Greylock, NVP
> ras ales
> 
> nae Ww
> ae 200M+ cep) isle] Ga
> 
> = V1 >i]
> 
> uw Microsoft =U

## 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.