Learn where FME meets AI in this upcoming webinar to offer you incredible time savings. This webinar is tailored to ignite imaginations and offer solutions to your data integration challenges. As the new digital era sets sail on the winds of AI, the tangibility of its integration in our daily schema is unfolding. Segment 1, titled “AI: The Good, the Bad and the FME” by Darren Fergus of Locus, navigates through the realms of AI, scrutinizing its pervasive impact while underscoring the symbiotic potential of FME and AI. Join in an engaging demonstration as FME and ChatGPT collaboratively orchestrate a PowerPoint narrative, epitomizing the alliance of AI with human ingenuity. In Segment 2, “Integrating GeoAI Models in FME” by Dennis Wilhelm and Dr. Christopher Britsch of con terra GmbH, the spotlight veers towards operationalizing AI in our daily tasks through FME. A practical approach to embedding GeoAI Models into FME Workspaces is unveiled, showcasing the ease of incorporating AI-driven methodologies into your FME workflows, skyrocketing productivity levels. To follow, Segment 3, "Unleash generative AI on your terms!" by Oliver Morris of Avineon-Tensing. While the prospects of Generative AI are thrilling, security and IT reservations, especially with 'phone home' tools, are genuine concerns. However, with open-source tools, you can locally harness large language models. In this demo, we'll unravel the magic of local AI deployment and its seamless integration into an FME workspace. Bonus! Dmitri will join us for a fourth segment to tie us off, showcasing what he has been up to this week, including using OpenAI API for texturing in FME, amoung other projects. Join us to explore the synergy of FME and AI: opening portals to a realm of revolutionized productivity and enriched user experiences.

1. Igniting Next Level
Productivity with
AI-Infused Data
Integration Workflows

2. Agenda
1 Introduction
2 Safe & FME
3 AI - the Good, the Bad and the FME by Locus
4 Integrating GeoAI Models in FME by con terra
5 Unlocking Generative AI on your own terms! by Tensing
6 Bonus! Segment by Dmitri
7 Conclusion & Resources
8 Next Steps
9 Q&A
Agenda

3. Welcome to Livestorm.
A few ways to engage with us during the webinar:
Audio issues? Click this for 4 simple
troubleshooting steps.

4. How to download slides
1. Hover over the
slide deck in the
webinar room
2. Click this button

5. 1
Introduction

6. AI
will take your job.
And give you one that’s better.
- Dharmesh Shah

7. 2
Safe & FME

8. 29+
27K+
128
190
20K+
years of solving data
challenges
FME Community
members
countries with
FME customers
organizations worldwide
global partners with
FME services
29+
29K+
128
140+
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

9. One platform, two technologies
FME Form FME Flow
Build and run data workflows Automate data workflows
FME Flow Hosted
Safe Software managed instance
fme.safe.com/platform
FME Enterprise Integration Platform
Safe & FME

10. 3
AI - the Good,
the Bad
and the FME

11. Darren
Fergus
Data Solutions
Locus

12. Introducing Locus
● Safe Software Platinum Partner
● Offices in New Zealand and Australia
● 15+ years on the FME journey; selling and supporting
● 7 experts with FME Technical Certifications
AI - the Good, the Bad and the FME

13. Agenda
1 Are We Now Doomed
2 Artificial Intelligence Defined
3 FME and AI
4 FME and IoT
5 FME and ChatGPT collaborate – the results

14. Introduction

15. We are all going to be
replaced by AI or are we?
Introduction

16. We are all going to be replaced by AI?!
1. Pick a subject?
2. Pick a presentation duration? – 1-5 Minutes
3. Pick the number of slides? – 1 to 6
4. Pick an emotion? – Funny, Scary Sad, Happy, Serious…
5. Pick a picture style
6. Off we go…
Introduction

17. Artificial Intelligence Defined
Instead, let’s ask AI, surely it should know?
Darren - “Define AI in 15 words exactly”
ChatGPT - "AI is the development of computer systems that simulate human intelligence and perform
intelligent tasks.“
Bing AI – “Artificial Intelligence (AI) is the ability of machines to perform tasks that typically require
human intelligence.”
Google Bard - “Artificial intelligence is the simulation of human intelligence in machines.”
Darren - “But that’s only 10 words?”
Google Bard - “You are right. Here is a definition of AI in 15 words exactly”
“The simulation of human intelligence in machines that are able to learn and act autonomously.”
Darren – “Thanks buddy”
Introduction

18. Artificial
Intelligence

19. Let’s meet some of the players
▪ There are many consumer systems out there, some are generic, others are very specific.
o ChatGPT – Google Bard – Bing AI
o AutoGPT
▪ Image Generation
o Dall-E-2
o Midjourney
o Leap.AI
▪ Others will become part of our working lives
o Microsoft 365 Copilot
Remember
me? - I’m
back!
Artificial Intelligence

20. FME and AI
▪ OpenAI ChatGPT Connector
▪ OpenAICompletions Connector
▪ OpenAIImageGenerator
▪ Leap.AI Connector
▪ HTTP Caller
o Google Bard
▪ Use AI to write a workspace
FME Haiku
Data is like clay
Mold it, shape it, transform it
Create something new
Form and Flow are one
Data is transformed, like clay
A new shape is born
Artificial Intelligence

21. Industry 4.0 (4 DOT ZERO)
▪ Industry 4.0 refers to the fourth industrial revolution, characterized by the
integration of advanced technologies and digitalization into various industrial
sectors
▪ The key features of Industry 4.0 include the following:
o Internet of Things (IoT)
o Big Data and Analytics
o Artificial Intelligence (AI)
o Automation and Robotics
o Cloud Computing
Artificial Intelligence

22. FME & IoT
▪ Azure IoT Connector
▪ Google IoT Core Connector
▪ AWS IoT Connector
▪ IBM IoT Connector
▪ MQTT Connector
▪ Kafka Apache
Artificial Intelligence

23. Conclusion
and Results

24. The Results | FME & ChatGPT Collaborate
● Here is the result
● Let’s take a look at the Workspace and what’s going on
Conclusion and Results

25. AI has emerged as one of the most
transformative technologies of our time and
has the potential to assist rather than replace
human endeavours, let’s embrace it!
Conclusion and Results

26. Takeaways
▪ There are more than 5,000 AI tools in the marketplace - if you haven’t explored some of
them, now is the time!
▪ People still matter but leveraging the power of these AI tools will improve
productivity, decision making and our innovation potential
Conclusion and Results

27. Thank
You!
Darren.Fergus@locus.co.nz

28. 4
Integrating
GeoAI Models
in FME

29. Dr. Christopher
Britsch
Consultant
con terra GmbH
Dennis
Wilhelm
Consultant
con terra GmbH

30. Agenda
1 What's GeoAI?
2 Integrating GeoAI in FME Processes
3 con terra GeoAI Framework
4 Project Example
5 Conclusion

31. What's
GeoAI?

32. GeoAI - Definition
GeoAI is a Machine Learning technology,
which enables the caption and analysis of
complex patterns and structures in
(geospatial) data.
What’s GeoAI?

33. Setup of a “typical” GeoAI Project
Data Preprocessing
AI-Methods
Integration and Operation
Visualization and Application
What’s GeoAI?

34. GeoAI at con terra
Data Enhancement
Natural Language Processing
Forecasting
Object and Image Recognition
GeoAI
What’s GeoAI?

35. Analyze existing
data
Accumulate
further data
Data
Preprocessing
Prepare first
PoC Model
Train and
Optimize Model
Integration
Monitoring and
Support
Use Case
Determination
What’s GeoAI?

36. Integrating where you need it, how you need it
What’s GeoAI?

37. Integrating
GeoAI
in FME

38. “Technology must be like
oxygen: ubiquitous,
necessary and invisible.”
- Chris Lehmann
Integrating GeoAI in FME

39. con terra
GeoAI
Framework

40. From Project to Product
● Next slides show our current work in progress
● Approach that has been used in multiple projects is now being generalized to ease usage
in the future
con terra GeoAI Framework

41. Main obstacles
● AI Model implementation is often performed by non-FME users in other environments
○ Jupyter Notebooks, Python Scripts, …
● Integration of these scripts into business processes if often unclear
● AI Models are usually based on many different libraries depending on the use case
○ Tensorflow, Keras, Pytorch, …
con terra GeoAI Framework

42. Main obstacles
● FME-users should be able to integrate these models
● Integration of these libraries in FME can be complicated (dependency mismatch, different
hardware requirements)
con terra GeoAI Framework

43. Our approach
● Provide an AI model runtime environment that can be maintained separate from FME
● Provide transformers to easily connect to this runtime from FME
con terra GeoAI Framework

44. Python environment
● Virtual Python environments for individual library management
○ Tensorflow, Pytorch, …
● Well defined interface to run models via HTTP or CLI
con terra GeoAI Framework

45. FME Process
con terra GeoAI Framework

46. New Transformer
con terra GeoAI Framework

47. Project
Example

48. Use Case: Anomaly Detection
• Various files are regularly uploaded to a system
○ Multiple dataset types with different schemas
○ Individual schemas consist of many columns
○ Defining individual rules for each attribute (or combination of attributes) would be
extremely time consuming
• Errors have occurred in previous uploads that compromise the system
Project Example

49. Solution
• We use an AI model to detect anomalies by comparing the current dataset to previous
datasets
○ Autoencoder using Tensorflow
○ Principal component analysis using sklearn
• Conditions
○ The schema for each dataset type is consistent
○ A minimum number of previous datasets is required
Project Example

50. What's an Autoencoder?
��
Input Bottleneck Output
🚀 ✅
Project Example

51. 🚗 ❌
Project Example
What's an Autoencoder?
��
Input Bottleneck Output

52. Project Example

53. Conclusion

54. Current Challenge - Integration
● Constructing GeoAI models becomes easier, however…
● …productively integration models can still be challenging
● Integration of models needs to be seamless…
● …in familiar and accustomed environments
Conclusion

55. Integrating GeoAI in FME
● The GeoAIConnector in FME solves the problem
○ Integration of pretrained models
○ Integration of custom made models
● Allows flexibility in your workspaces
○ There isn't one model for everything ➡ Flexibility to include custom implementations
necessary
● Not every functionality has to be implemented in FME, but everything can be connected!
Conclusion

56. c.britsch@conterra.de
d.wilhelm@conterra.de
Thank
You!

57. 6
Unlocking the
power of
generative AI on
your own terms!

58. Oliver Morris
Business Director
Tensing

59. Agenda
● Intro - hype to reality
● It’s not just chatGPT (and why that matters)
● Privacy and security concerns
● Integrating locally running, open-source LLM’s
● Sneak peek - multi-modal models
● Wrap up

61. Generative AI ≠ chatGPT
Platforms including APIs
● OpenAI, Bard, Midjourney
Models
● gpt-3.5-turbo, gpt-4, PaLM2
● Llama, Mistral, LLaVA

62. Privacy / Security

63. https://www.gov.uk/government/publications/guidance-to-civil-servants-on-u
se-of-generative-ai/guidance-to-civil-servants-on-use-of-generative-ai
Government Guidelines

64. Cost

65. Locally deploy Generative AI tools
Ollama.ai - open source, permissive MIT license
● Terminal and API interface, runs on Linux, Mac and Windows*
Mistral.ai
● requires 5GB drive space and needs approx. 8GB RAM
Security / cost does not need to be a blocker for using Generative AI

66. Step 1 - Install Windows Subsystem for Linux (WSL2)

67. Step 2 - install Ubuntu/Debian

68. Step 3 - install Ollama

69. Step 4 - install/run and talk to a LLM

70. Generate a code-reviewer
Ollama installed,
mistral model
enabled
Time to create a
prompt…

71. Ollama has an API
- bring on
HTTPCaller

72. FME makes
things easier
Tensing verified publisher
FME Hub Transformer
Orchestration tool
Plug and play
bit.ly/FMELocalGenAI

73. Multi-modal models
https://www.researchgate.net/figure/Generative-AI-models-Unimodal-and-multi-modal-examples_fig1_369771657

74. FME makes
things easier

75. Visual AI - LLaVA vs ChatGPT4 Vision

76. Wrap up
You can use Generative AI, keep your IT dept at bay and you budget in check
Tasks such as data cleanup, SQL and Python reviewing are easy to integrate into
your traditional data ETL pipelines
Ollama and FME can help you get started in this new domain
Be patient, it’s only going to get better

77. omorris@tensing.com
Thank
You!

78. Bonus!
Segment
with Dmitri

79. 7
Resources

80. Get our Ebook
Spatial Data for the
Enterprise
fme.ly/gzc
Guided learning experiences
at your fingertips
community.safe.com
/s/academy
FME Academy

81. Check out how-to’s & demos
in the knowledge base
community.safe.com
/s/knowledge-base
Knowledge Base Webinars
Upcoming & on-demand
webinars
safe.com/webinars

82. 8
Next Steps

83. We’d love to help you get
started.
Get in touch with us at
info@safe.com
Experience the FME Accelerator
Contact Us
Unlock the power of your
data in only 90 minutes
Register for free at
fme.safe.com/accelerator

84. ClaimYour Community Badge
● Get community badges for watching
webinars!
● fme.ly/WebinarBadge
● Today’s code: MSLGF
Join the Community today!

85. 9
Q&A

86. ThankYou
Recap of Next Steps
1 Join the FME Community
2 Contact today’s guest speakers:
Darren.Fergus@locus.co.nz
c.britsch@conterra.de
d.wilhelm@conterra.de
omorris@tensing.com
3 Experience the FME Accelerator
Please fill out our
webinar survey