AI Agents Made Simple:
Unleash the Power of AllYour
Data with Any AI
Dmitri
Bagh
Scenarios
Chris
Berger
Customer Solutions
Team Lead
Don
Murray
CEO
Meet the Team
Siobhan
Ryan
Technical Specialist
Avineon-Tensing
Welcome to Livestorm.
A few ways to engage with us during the webinar:
Audio issues? Click this for 4 simple
troubleshooting steps.
How to download slides
1. Hover over the
slide deck in the
webinar room
2. Click this button
● May 6-8, 2025
● Be among the first to hear ground breaking
updates and launches!
● *New* - Get FME certified for free at the event!
○ 2 exams offered on Wednesday, May 7.
Poll
How familiar are you with
AI Agents?
Poll
Which AI services have you
worked with?
(Select all that apply)
Agenda
1 Introduction
2 All Data, Any-AI Workflows
3 Real-Time Automation in Action
4 Dynamic Parameters: A Sneak Peek
5 Avineon-Tensing: Agentic Workflow Flash Demo
6 Empowering AR with AI
7 Conclusion
8 Resources & Next Steps
9 Q&A
Agenda
1
Introduction
● Autonomous and intelligent.
● Perform tasks like
data quality, data extraction,
decision support and more.
Often in real-time.
● Combining data processing with
AI brings more power to
automated workflows.
What are AI Agents?
Future of AI in Data
Pipelines
AI and AI Agents enhance automation
and scalability.
Join us at the Peak of Data and AI to
learn more!
Data Integration and AI unlocks new
possibilities.
All-data coupled with Any-AI makes
pipelines faster, smarter, and more
intuitive.
What’s Next?
AI Agents and Real-Time data quality is theme today.
What you will see today
● AI driven schema mapping to enhance quality of data mapping.
● Real-Time meter using photos to reduce costs and increases accuracy.
● AR and AI combine to bring the power of AI to the field worker.
● AI performing QA of geospatial data. AI is not just for standard attributes.
2
All-Data,
Any-AI
Workflows
Only FME brings All-Data to Any-AI
GIS
CAD
Database
XML
Raster
3D
BIM
Web
Point
Cloud
Cloud
Big Data
IOT
Gaming
BI
Indoor
Mapping
AR/VR
Generative
AI
Cloud
Native
Tabular
Graph
Any AI cloud provider, Any locally hosted AI Model
Pre-built connectors available (Amazon, Google, Ollama, OpenAI)
Discover more on the FME Hub!
Integrate Cloud Based AI Services
Integrate Embedded AI Services
● Build a framework to exploit LLM models locally.
● Compliment the cloud based services.
● Data stays local reducing any security concerns.
FME & AI Services
3
Real-Time
Automation
in Action
Agentic Schema
Mapping
● Leverage AI for intelligent parsing
and structuring of highly variable
address data.
● FME’s dynamic workspace
capabilities enable seamless
integration.
● Outputs are standardized, ready for
downstream use.
AI Assisted Data Integration
Agentic Schema
Mapping
● Leverage AI for intelligent parsing
and structuring of highly variable
address data.
● FME’s dynamic workspace
capabilities enable seamless
integration.
● Outputs are standardized, ready for
downstream use.
AI Assisted Data Integration
Agentic Schema
Mapping
● Leverage AI for intelligent parsing
and structuring of highly variable
address data.
● FME’s dynamic workspace
capabilities enable seamless
integration.
● Outputs are standardized, ready for
downstream use.
AI Assisted Data Integration
Schema Variability
Challenge
● Data providers deliver address
datasets with inconsistent schemas,
data types, and in various data
formats.
● Variability makes it difficult to
integrate data into downstream
processes.
● Manual efforts to standardize data
are time-consuming and prone to
errors.
AI Assisted Data Integration
Schema Variability
Solution
● AI Processing can be used to:
○ Identify key attributes (e.g., street
name, number, postal code).
○ Concatenate or parse fields
dynamically.
○ Resolve inconsistencies with
Structured Outputs
● Output: Standardized attributes for
seamless downstream use
AI Assisted Data Integration
Schema Variability
Solution
● AI Processing can be used to:
○ Identify key attributes (e.g., street
name, number, postal code).
○ Concatenate or parse fields
dynamically.
○ Resolve inconsistencies with
Structured Outputs
● Output: Standardized attributes for
seamless downstream use
AI Assisted Data Integration
Slide Title
Transform
provided data
into standard
schema.
Goal Block Key
Agentic Data Integration
Result
Inconsistent
schemas and
various data
formats
Leverage AI
Agents using
Structured
Outputs
Standardized
schema that is
ready for
downstream
processes
Demo
Key Benefits
● Efficiency: Dramatically reduces
processing & QA time.
● Accuracy: Handles complex data
ambiguities with precision.
● Scalability: Processes large
volumes of data with ease.
● Cost-Effective: Reduces manual
intervention and associated costs.
Why Use AI for Schema Mapping?
● Download the latest AI Transformers
○ GoogleGeminiConnector (V3)
○ OpenAIChatGPTConnector (V7)
● Getting Started with JSON
● Vertex Studio & ChatGPT
○ Output Format Examples
○ Use the chat playground to generate
the required JSON
Leverage AI Agents in your
Workflows
Where to Start
4
Dynamic
Parameters: A
Sneak Peek
Published Parameters
Prompts user for input at runtime
● Choice, color, datetime, file/URL, etc
● Defined in the .fmw file (authored in
FME Workbench)
See Working with User Parameters for more information
Workspace Apps and Published Parameters
Dynamic Parameters
Dynamically Updating Published
Parameter Values
● Updates via Workspace, local file,
remote file, etc
Workspace Apps and Published Parameters
Dynamic Parameters
Dynamically Updating Published
Parameter Values
● Updates via Workspace, local file,
remote file, etc
● JSON based
○ Structured Outputs make it easy
to populate display, value, and
default value
Workspace Apps and Published Parameters
User Guided Workflows
Similar to the fully Automated process
described in section 4, this provides
organizations with a front end or
intermediate visual confirmation step.
● Publicly accessible or authenticated
● Makes data integration more
accessible
AI Assisted Data Integration
Slide Title
Enable smarter,
faster geocoding
setup.
Goal Block Key
Geocoding with FME Flow Workspace Apps
Result
Inconsistent
schemas slow
down user
workflows.
AI agents and
Dynamic
Parameters
automate schema
mapping.
Accelerate
processes and
create an
intuitive user
experience.
Demo
AI Enhanced Data Integration
AI Enhanced Data Integration
● Dynamic Parameters Simplify
Workflows… coming soon
● AI Enhances Usability
● Multi-step Workspace apps prevent
the need to custom code
● Seamless User Experience
Accelerate Workflows with AI
AI Enhanced Data Integration
5
Agentic Workflow
Flash Demo
Article | Push Data from ArcGIS Survey123 to an Application in Real-time with Webhooks: fme.ly/Survey123ArcGISW
Demo
Sending alerts
● TwilioSMSSender
Uses the Twilio API to allow you to send SMS,
WhatsApp messages, calls from FME
Workspaces.
● TelegramConnector
Sends messages (attribute text or images) to
chats using a Telegram Bot.
Free FME AI Skills Booster
4th March
11am / 5pm GMT
12pm / 6pm CET
6am / 12pm EST
3am / 9am PST
tensing.com/en/fme-training
Siobhan Ryan
Technical Specialist
sryan@avineon-tensing.com
Thank you
6
Empowering
AR with AI
AI Agents
● Field Notes to Reports
transforms field audio notes into structured reports, interpreting intent, enhancing
details with feature attributes, and ensuring clear, professional documentation.
● Smart QA analysis
validates attributes and geometry not in isolation, but in the full context of all related
data.
AI Agents
Bring in an object in any
format
(GIS, CAD, DB, XML, etc.)
Dynamically transform
attributes and geometry
to GeoJSON
Write QA instructions,
include GeoJSON
Send instructions to AI, get
structured JSON, HTML, etc. back
Generate and stream
HTML
AI Agents
Bring in an object in any
format
(GIS, CAD, DB, XML, etc.)
Dynamically transform
attributes and geometry
to GeoJSON
Write QA instructions,
include GeoJSON
Send instructions to AI, get
structured JSON, HTML, etc. back
Generate and stream
HTML
AI Agent: Field Notes to Reports
A static page where AI
transforms audio input
into text
AI Agent: Field Notes to Reports
AI agent prepares a
report, which includes
professional languages
and all the details about
the object.
AI Agent: Smart QA Analysis
{
"issues": [
{
"id": "SNMH01022",
"feature_type": "SewerMain",
"attribute": "Status",
"value": "Operatng",
"description": "Typo in the word Operatng - must be Operating",
"severity": "medium"
},
{
"id": "SNMH01022",
"feature_type": "SewerMain",
"attribute": "Depth",
"value": "2.5m",
"description": "Unit name 'm' is used in the attribute representing
Depth",
"severity": "medium"
},
{
"id": "SNMH01022",
"feature_type": "SewerMain",
"attribute": "FlowCapacity",
"value": "1500 m³/day",
"description": "Unit name 'm³/day' is used in the attribute
representing FlowCapacity",
"severity": "medium"
},
{
"id": "SNMH01022",
"feature_type": "SewerMain",
"attribute": "PressureRating",
"value": "4 bar",
"description": "Unit name 'bar' is used in the attribute
representing PressureRating",
"severity": "medium"
},
{
"id": "SNMH01022",
"feature_type": "SewerMain",
"attribute": "Grade",
"value": 0.5,
"description": "The calculated grade is 0.5, which matches the
existing grade attribute. No issues found.",
"severity": "low"
}
]
}
Notice the geometry QA
Poll:
Would you attend a training course to take a
deeper-dive into applying AI Agents to your
own workflows?
7
Conclusion
Summary
AI Agents can make a big impact on Real-Time QA.
FME is the only All-Data,
Any-AI Platform.
Easily create AI Agents.
Without writing a single line
of code.
FME Flow Hosted
SaaS version of FME Flow
The only All-Data, Any-AI Platform
FME Form FME Flow
The Authoring environment for all workflows. The Deployment environment delivering services to
your organization.
fme.safe.com/platform
FME Enterprise Integration Platform
29+
27K+
128
190
20K+
years of solving data
challenges
FME Community
members
countries with
FME customers
organizations worldwide
global partners with
FME services
30+
37K+
128
170+
25K+
years of solving data
challenges
FME Community
members
countries with
FME customers
organizations worldwide
global partners with
FME services
200K+
users worldwide
Safe & FME
8
Resources &
Next Steps
Get our Ebook
Spatial Data for the
Enterprise
fme.ly/gzc
Guided learning
experiences at your
fingertips
academy.safe.com
FME Academy
Resources
Check out how-to’s &
demos in the knowledge
base
support.safe.com
Knowledge Base Webinars
Upcoming &
on-demand webinars
safe.com/webinars
Check out
our podcasts
on-demand.
featuring special guest
speakers over at EM360
Resources
We’d love to help you get
started.
Get in touch with us at
info@safe.com
Experience the
FME Accelerator
Contact Us
A world where data is not just a
commodity but a catalyst for
real change.
fme.safe.com/accelerator
Next Steps
ClaimYour Community Badge &
Dive into the new Community!
● Get community badges for watching
webinars
● community.safe.com
● Today’s code: N9HW1A
Join the Community today!
Next Steps
● May 6-8, 2025
● Be among the first to hear ground breaking
updates and launches!
● *New* - Get FME certified for free at the event!
○ 2 exams offered on Wednesday, May 7.
9
Q&A
ThankYou
Recap of Next Steps
1 Follow us on LinkedIn!
2 Contact us
3 Experience the FME Accelerator
Please fill out our
webinar survey
Streamline workflows,
integrate seamlessly with all
data, any AI, and automate
decision-making in real-time.

AI Agents Made Simple: Unleash the Power of All Your Data with Any AI

  • 1.
    AI Agents MadeSimple: Unleash the Power of AllYour Data with Any AI
  • 2.
    Dmitri Bagh Scenarios Chris Berger Customer Solutions Team Lead Don Murray CEO Meetthe Team Siobhan Ryan Technical Specialist Avineon-Tensing
  • 3.
    Welcome to Livestorm. Afew ways to engage with us during the webinar: Audio issues? Click this for 4 simple troubleshooting steps.
  • 4.
    How to downloadslides 1. Hover over the slide deck in the webinar room 2. Click this button
  • 5.
    ● May 6-8,2025 ● Be among the first to hear ground breaking updates and launches! ● *New* - Get FME certified for free at the event! ○ 2 exams offered on Wednesday, May 7.
  • 6.
    Poll How familiar areyou with AI Agents?
  • 7.
    Poll Which AI serviceshave you worked with? (Select all that apply)
  • 8.
    Agenda 1 Introduction 2 AllData, Any-AI Workflows 3 Real-Time Automation in Action 4 Dynamic Parameters: A Sneak Peek 5 Avineon-Tensing: Agentic Workflow Flash Demo 6 Empowering AR with AI 7 Conclusion 8 Resources & Next Steps 9 Q&A Agenda
  • 9.
  • 10.
    ● Autonomous andintelligent. ● Perform tasks like data quality, data extraction, decision support and more. Often in real-time. ● Combining data processing with AI brings more power to automated workflows. What are AI Agents?
  • 11.
    Future of AIin Data Pipelines AI and AI Agents enhance automation and scalability. Join us at the Peak of Data and AI to learn more! Data Integration and AI unlocks new possibilities. All-data coupled with Any-AI makes pipelines faster, smarter, and more intuitive. What’s Next?
  • 12.
    AI Agents andReal-Time data quality is theme today. What you will see today ● AI driven schema mapping to enhance quality of data mapping. ● Real-Time meter using photos to reduce costs and increases accuracy. ● AR and AI combine to bring the power of AI to the field worker. ● AI performing QA of geospatial data. AI is not just for standard attributes.
  • 13.
  • 14.
    Only FME bringsAll-Data to Any-AI GIS CAD Database XML Raster 3D BIM Web Point Cloud Cloud Big Data IOT Gaming BI Indoor Mapping AR/VR Generative AI Cloud Native Tabular Graph
  • 15.
    Any AI cloudprovider, Any locally hosted AI Model Pre-built connectors available (Amazon, Google, Ollama, OpenAI) Discover more on the FME Hub! Integrate Cloud Based AI Services
  • 16.
    Integrate Embedded AIServices ● Build a framework to exploit LLM models locally. ● Compliment the cloud based services. ● Data stays local reducing any security concerns. FME & AI Services
  • 17.
  • 18.
    Agentic Schema Mapping ● LeverageAI for intelligent parsing and structuring of highly variable address data. ● FME’s dynamic workspace capabilities enable seamless integration. ● Outputs are standardized, ready for downstream use. AI Assisted Data Integration
  • 19.
    Agentic Schema Mapping ● LeverageAI for intelligent parsing and structuring of highly variable address data. ● FME’s dynamic workspace capabilities enable seamless integration. ● Outputs are standardized, ready for downstream use. AI Assisted Data Integration
  • 20.
    Agentic Schema Mapping ● LeverageAI for intelligent parsing and structuring of highly variable address data. ● FME’s dynamic workspace capabilities enable seamless integration. ● Outputs are standardized, ready for downstream use. AI Assisted Data Integration
  • 21.
    Schema Variability Challenge ● Dataproviders deliver address datasets with inconsistent schemas, data types, and in various data formats. ● Variability makes it difficult to integrate data into downstream processes. ● Manual efforts to standardize data are time-consuming and prone to errors. AI Assisted Data Integration
  • 22.
    Schema Variability Solution ● AIProcessing can be used to: ○ Identify key attributes (e.g., street name, number, postal code). ○ Concatenate or parse fields dynamically. ○ Resolve inconsistencies with Structured Outputs ● Output: Standardized attributes for seamless downstream use AI Assisted Data Integration
  • 23.
    Schema Variability Solution ● AIProcessing can be used to: ○ Identify key attributes (e.g., street name, number, postal code). ○ Concatenate or parse fields dynamically. ○ Resolve inconsistencies with Structured Outputs ● Output: Standardized attributes for seamless downstream use AI Assisted Data Integration
  • 24.
    Slide Title Transform provided data intostandard schema. Goal Block Key Agentic Data Integration Result Inconsistent schemas and various data formats Leverage AI Agents using Structured Outputs Standardized schema that is ready for downstream processes
  • 25.
  • 26.
    Key Benefits ● Efficiency:Dramatically reduces processing & QA time. ● Accuracy: Handles complex data ambiguities with precision. ● Scalability: Processes large volumes of data with ease. ● Cost-Effective: Reduces manual intervention and associated costs. Why Use AI for Schema Mapping?
  • 27.
    ● Download thelatest AI Transformers ○ GoogleGeminiConnector (V3) ○ OpenAIChatGPTConnector (V7) ● Getting Started with JSON ● Vertex Studio & ChatGPT ○ Output Format Examples ○ Use the chat playground to generate the required JSON Leverage AI Agents in your Workflows Where to Start
  • 28.
  • 29.
    Published Parameters Prompts userfor input at runtime ● Choice, color, datetime, file/URL, etc ● Defined in the .fmw file (authored in FME Workbench) See Working with User Parameters for more information Workspace Apps and Published Parameters
  • 30.
    Dynamic Parameters Dynamically UpdatingPublished Parameter Values ● Updates via Workspace, local file, remote file, etc Workspace Apps and Published Parameters
  • 31.
    Dynamic Parameters Dynamically UpdatingPublished Parameter Values ● Updates via Workspace, local file, remote file, etc ● JSON based ○ Structured Outputs make it easy to populate display, value, and default value Workspace Apps and Published Parameters
  • 32.
    User Guided Workflows Similarto the fully Automated process described in section 4, this provides organizations with a front end or intermediate visual confirmation step. ● Publicly accessible or authenticated ● Makes data integration more accessible AI Assisted Data Integration
  • 33.
    Slide Title Enable smarter, fastergeocoding setup. Goal Block Key Geocoding with FME Flow Workspace Apps Result Inconsistent schemas slow down user workflows. AI agents and Dynamic Parameters automate schema mapping. Accelerate processes and create an intuitive user experience.
  • 34.
  • 35.
    AI Enhanced DataIntegration
  • 36.
    AI Enhanced DataIntegration
  • 37.
    ● Dynamic ParametersSimplify Workflows… coming soon ● AI Enhances Usability ● Multi-step Workspace apps prevent the need to custom code ● Seamless User Experience Accelerate Workflows with AI AI Enhanced Data Integration
  • 38.
  • 41.
    Article | PushData from ArcGIS Survey123 to an Application in Real-time with Webhooks: fme.ly/Survey123ArcGISW
  • 43.
  • 44.
    Sending alerts ● TwilioSMSSender Usesthe Twilio API to allow you to send SMS, WhatsApp messages, calls from FME Workspaces. ● TelegramConnector Sends messages (attribute text or images) to chats using a Telegram Bot.
  • 45.
    Free FME AISkills Booster 4th March 11am / 5pm GMT 12pm / 6pm CET 6am / 12pm EST 3am / 9am PST tensing.com/en/fme-training
  • 46.
  • 47.
  • 48.
    AI Agents ● FieldNotes to Reports transforms field audio notes into structured reports, interpreting intent, enhancing details with feature attributes, and ensuring clear, professional documentation. ● Smart QA analysis validates attributes and geometry not in isolation, but in the full context of all related data.
  • 49.
    AI Agents Bring inan object in any format (GIS, CAD, DB, XML, etc.) Dynamically transform attributes and geometry to GeoJSON Write QA instructions, include GeoJSON Send instructions to AI, get structured JSON, HTML, etc. back Generate and stream HTML
  • 50.
    AI Agents Bring inan object in any format (GIS, CAD, DB, XML, etc.) Dynamically transform attributes and geometry to GeoJSON Write QA instructions, include GeoJSON Send instructions to AI, get structured JSON, HTML, etc. back Generate and stream HTML
  • 51.
    AI Agent: FieldNotes to Reports A static page where AI transforms audio input into text
  • 52.
    AI Agent: FieldNotes to Reports AI agent prepares a report, which includes professional languages and all the details about the object.
  • 53.
    AI Agent: SmartQA Analysis { "issues": [ { "id": "SNMH01022", "feature_type": "SewerMain", "attribute": "Status", "value": "Operatng", "description": "Typo in the word Operatng - must be Operating", "severity": "medium" }, { "id": "SNMH01022", "feature_type": "SewerMain", "attribute": "Depth", "value": "2.5m", "description": "Unit name 'm' is used in the attribute representing Depth", "severity": "medium" }, { "id": "SNMH01022", "feature_type": "SewerMain", "attribute": "FlowCapacity", "value": "1500 m³/day", "description": "Unit name 'm³/day' is used in the attribute representing FlowCapacity", "severity": "medium" }, { "id": "SNMH01022", "feature_type": "SewerMain", "attribute": "PressureRating", "value": "4 bar", "description": "Unit name 'bar' is used in the attribute representing PressureRating", "severity": "medium" }, { "id": "SNMH01022", "feature_type": "SewerMain", "attribute": "Grade", "value": 0.5, "description": "The calculated grade is 0.5, which matches the existing grade attribute. No issues found.", "severity": "low" } ] } Notice the geometry QA
  • 54.
    Poll: Would you attenda training course to take a deeper-dive into applying AI Agents to your own workflows?
  • 55.
  • 56.
    Summary AI Agents canmake a big impact on Real-Time QA.
  • 57.
    FME is theonly All-Data, Any-AI Platform.
  • 58.
    Easily create AIAgents. Without writing a single line of code.
  • 59.
    FME Flow Hosted SaaSversion of FME Flow The only All-Data, Any-AI Platform FME Form FME Flow The Authoring environment for all workflows. The Deployment environment delivering services to your organization. fme.safe.com/platform FME Enterprise Integration Platform
  • 60.
    29+ 27K+ 128 190 20K+ years of solvingdata challenges FME Community members countries with FME customers organizations worldwide global partners with FME services 30+ 37K+ 128 170+ 25K+ years of solving data challenges FME Community members countries with FME customers organizations worldwide global partners with FME services 200K+ users worldwide Safe & FME
  • 61.
  • 62.
    Get our Ebook SpatialData for the Enterprise fme.ly/gzc Guided learning experiences at your fingertips academy.safe.com FME Academy Resources Check out how-to’s & demos in the knowledge base support.safe.com Knowledge Base Webinars Upcoming & on-demand webinars safe.com/webinars
  • 63.
    Check out our podcasts on-demand. featuringspecial guest speakers over at EM360 Resources
  • 64.
    We’d love tohelp you get started. Get in touch with us at info@safe.com Experience the FME Accelerator Contact Us A world where data is not just a commodity but a catalyst for real change. fme.safe.com/accelerator Next Steps
  • 65.
    ClaimYour Community Badge& Dive into the new Community! ● Get community badges for watching webinars ● community.safe.com ● Today’s code: N9HW1A Join the Community today! Next Steps
  • 66.
    ● May 6-8,2025 ● Be among the first to hear ground breaking updates and launches! ● *New* - Get FME certified for free at the event! ○ 2 exams offered on Wednesday, May 7.
  • 67.
  • 68.
    ThankYou Recap of NextSteps 1 Follow us on LinkedIn! 2 Contact us 3 Experience the FME Accelerator Please fill out our webinar survey
  • 69.
    Streamline workflows, integrate seamlesslywith all data, any AI, and automate decision-making in real-time.