AI Agents with Gemini 2.0
Beyond the Chatbot
Márton Kodok
Google Developer Expert at REEA.net
#geecon | Room 12
1. Gemini - The Big Picture
2. Gemini - capabilities
3. Enabling the agentic era
4. Demos
5. Conclusions
Agenda
AI Agents with Gemini 2.0 - Beyond the Chatbot @martonkodok
Márton Kodok
AI Agents with Gemini 2.0 - Beyond the Chatbot @martonkodok
● 8x Google Developer Expert on Cloud technologies (2016-->)
● 3x Champion of Google Cloud Innovators program (2021-->)
● Among the Top 3 romanians on Stackoverflow 208k reputation
● Crafting Cloud Architecture+ML backends at REEA.net
Articles: martonkodok.medium.com
Twitter: @martonkodok
Slideshare:martonkodok
StackOverflow: pentium10
GitHub: pentium10
What LLMs are, and are not
AI Agents with Gemini 2.0 - Beyond the Chatbot @martonkodok
“Large Language
Models are not a
source of truth.”
– Noble Ackerson
What they are,
however, is a way to
transform one thing
into another.
● Phrases into code
fragments
● Structured data
into textual results
Combine them with
APIs that you already
know and work with,
and your applications
can have amazing
new capabilities.
A powerful new tool!
Gemini - The Big Picture
AI Agents with Gemini 2.0 - Beyond the Chatbot @martonkodok
Gemini Models
ML Other Models
Client Libs & SDK
Gemini API
Vertex AI
AI Studio
Local
Models
Gemma
ADK
AI Tools for
Business
AI assistant
Gemini for
Workspace
Gemini
chat app
Gemini Live
Notebook LM
Developers
"Consumers"
✅ ✅
✅
✅
✅
@martonkodok
Gemini comes in different sizes
2.5 Pro
2.5 Flash
2.0 Pro
1.5 Flash
✅
❌
Multimodal Input
(text, img, video, audio, PDF)
Context Window + Output
Native Tool Use
(Search, Code Exec)
Thinking Mode
1M / 8K
❌ ❌
✅ ✅ ✅
2M / 8K 1M / 65K 1M / 65K
set Thinking Budget
10K free Search queries
Preview
Preview
AI Agents with Gemini 2.0 - Beyond the Chatbot @martonkodok
Texttoimagegeneration
withhighqualityandspeed
● Aspectratios
1:1, 9:16, 16:9, 3:4, 4:3
● Styles
photorealism, paintings, richly
textured, oil paintings, comic
● UseCases
Banners, Posters, Advertizing
A lone skateboarder glides through the echoing stillness of
an abandoned train station, bathed in the soft glow of a
single overhead lamp, creating a dramatic silhouette
against the vast, empty space.
A dynamic, high-energy animation of a superhero flying
across a busy metropolis with a blurred background
A watercolor painting of mt fuji A bright style- magazine shot of a woman putting on
mascara.
Imagen 3
Demo
Gemini: Native image generation
AI Agents with Gemini 2.0 - Beyond the Chatbot @martonkodok
AI Agents with Gemini 2.0 - Beyond the Chatbot @martonkodok
Veo 2
Veo2
Vision of abandoned Warsaw, capital of Poland @Tomek Porozynski
Veo 2
● Author: Tomek Porozynski
● GDE & Cloud Architect
● Made with Veo 2
AI Agents with Gemini 2.0 - Beyond the Chatbot @martonkodok
● Local model runs on 1 single GPU
● MultiLingual 140+ languages
● Function Calling
● 1b - 4b - 12b - 27b sizes
● Vision - interleaved image & text input
Gemma 3
Getting started with the
Gemini API
AI Agents with Gemini 2.0 - Beyond the Chatbot @martonkodok
AI Agents with Gemini 2.0 - Beyond the Chatbot @martonkodok
aistudio.google.com
AI Studio
➔ Generate API key
➔ Create, test, and save prompts
➔ SDK in Python/Go/JS+TS/Firebase
➔ Java via LangChain4j library
AI Agents with Gemini 2.0 - Beyond the Chatbot @martonkodok
AI Studio
➔ UI is free up to 5-10 RPM
➔ Prompt management
➔ Access to Gemma 3.
➔ Unified SDK references
AI Studio mostly as first
example
AI Agents with Gemini 2.0 - Beyond the Chatbot @martonkodok
console.cloud.google.com/vertex-ai/studio
Vertex AI Studio
➔ Service account key
➔ Vertex AI SDK
➔ Access to 120 models via Model Garden
➔ Available for individual/edu/business
➔ Java via google-cloud-vertexai Library
AI Agents with Gemini 2.0 - Beyond the Chatbot @martonkodok
console.cloud.google.com/vertex-ai/studio
Vertex AI Studio
➔ Project based use
➔ Access to preview models/increased quotas
➔ Help me write prompts
➔ Imagen 3 + Veo 2
➔ Prompt management
AI Agents with Gemini 2.0 - Beyond the Chatbot @martonkodok
gemini.google.com
aistudio.google.com
ai.google.dev/gemini-api
console.cloud.google.com/vertex-ai/studio
Consumers Enterprise
Developers
Unified SDK
AI Agents with Gemini 2.0 - Beyond the Chatbot @martonkodok
Gemini
Capabilities
AI Agents with Gemini 2.0 - Beyond the Chatbot @martonkodok
What makes Gemini unique?
AI Agents with Gemini 2.0 - Beyond the Chatbot @martonkodok
Native
Tool Use
Multimodal
in and out
Long context length
Context Caching
Multimodal
Live API
Thinking Mode
Budget driven
Spatial
Understanding
AI Agents with Gemini 2.0 - Beyond the Chatbot @martonkodok
Function Calling
API parameters Tools Execution
Processing
Model Output
Function Calls Function Responses
Data
User
Prompt
Generating
Live Info
AI Agents with Gemini 2.0 - Beyond the Chatbot @martonkodok
Google Search Tool
google_search_tool = Tool(google_search=GoogleSearch())
response = client.models.generate_content(
model="gemini-2.0-flash-001",
contents="How’s the weather like today in London?",
config=GenerateContentConfig(
tools=[google_search_tool]
))
AI Agents with Gemini 2.0 - Beyond the Chatbot @martonkodok
Code Execution Tool
code_execution_tool = Tool(code_execution=ToolCodeExecution())
response = client.models.generate_content(
model="gemini-2.0-flash-001",
contents="What is the sum of the first 50 prime numbers?",
config=GenerateContentConfig(
tools=[code_execution_tool],
temperature=0))
AI Agents with Gemini 2.0 - Beyond the Chatbot @martonkodok
Spatial Understanding
aistudio.google.com/starter-apps/spatial
➔ 2D and 3D spatial understanding
➔ Segmentations masks
➔ Points
➔ 2D and 3D bounding boxes
➔ Local modal alternative: PaliGemma 2
Gemini 2.0
Enabling the agentic era
AI Agents with Gemini 2.0 - Beyond the Chatbot @martonkodok
AI Agents with Gemini 2.0 - Beyond the Chatbot @martonkodok
Core Pillars of ADK
➔ Open Source
➔ Code agents like Google
➔ The only dependency is a model
➔ Run on your laptop deploy anywhere
➔ Orchestrate automatically in sequence/parallel
➔ OpenAPI tools
➔ Supports Toolbox, MCPs, A2A
Agent Development Kit
AI Agents with Gemini 2.0 - Beyond the Chatbot @martonkodok
Build multi-agent capabilities
➔ Function tools
➔ Built-in tools
➔ Third party tools
➔ Agent Garden on Google Cloud
➔ MCP Tools
➔ OpenAPI tools
➔ Agent2Agent Protocol (A2A)
AI Agents with Gemini 2.0 - Beyond the Chatbot @martonkodok
ADK vs Langgraph vs LLamaindex
➔ ADK has Gemini and LiteLLM support
Langgraph integrates with Langchain
➔ Agent workflow and delegation:
ADK uses LLM to decide agent routes
Langgraph controlled by stategraph/nodes
➔ Both support short- and long-term memory
ADK uses output key to manage state for multiple
agents - feels better.
➔ Guardrails:
ADK uses callback before/post LLM validation, which
can be more centralized control/cleaner
Langgraph needs you to manipulate state/LCEL
➔ Llamaindex focused on RAG knowledge assistants Multi-agent documentation from LangGraph.
AI Agents with Gemini 2.0 - Beyond the Chatbot @martonkodok
Flexibility in the way you interact with your agents
Demo 1
ADK agent
AI Agents with Gemini 2.0 - Beyond the Chatbot @martonkodok
Agent Development Kit
Gemini 2.0 use case
🍷wine collection into structured data
AI Agents with Gemini 2.0 - Beyond the Chatbot @martonkodok
AI Agents with Gemini 2.0 - Beyond the Chatbot @martonkodok
Dataset
➔ Thousands of professional photos
➔ Front + back side of the bottle(s)
➔ High detail photo quality
AI Agents with Gemini 2.0 - Beyond the Chatbot @martonkodok
Challenges with Gemini 2.0
➔ High quality dataset - clear text is visible on zoom
➔ Initial dry run produced ZERO success.
What could be the problem?
➔ LLMs works best with rectangle images
➔ Image greater > 384 pixels -> is cropped into tiles...
➔ Each tile is resized to 768x768
➔ and uses 258 tokens
AI Agents with Gemini 2.0 - Beyond the Chatbot @martonkodok
media_resolution config
➔ To control how input images are processed,
use the `media_resolution` config parameter:
➔ MEDIA_RESOLUTION_LOW:
the image is tokenized into 66 tokens (fixed).
➔ MEDIA_RESOLUTION_MEDIUM:
the image is tokenized into 258 tokens (fixed).
➔ MEDIA_RESOLUTION_HIGH:
the input image is split (if needed) into non-overlapping
crops, leading to X tiles of 258 tokens.
Gemini Spatial Understanding
returning segmentation masks
AI Agents with Gemini 2.0 - Beyond the Chatbot @martonkodok
AI Agents with Gemini 2.0 - Beyond the Chatbot @martonkodok
Gemini Spatial Understanding
➔ 2D bounding boxes
➔ Segmentation masks
➔ Points
AI Agents with Gemini 2.0 - Beyond the Chatbot @martonkodok
Extract info from images
You are an expert in text extraction
from images. Your task is to extract
all text from the wine bottle
label(s) in the **attached images**.
Combine the text from both images.
Return the text in markdown format,
preserving any formatting (e.g.,
bold, italics, lists) as closely as
possible.
Use headings (e.g., #, ##) to
organize the text if appropriate.
Using Gemini 2.0
AI Agents with Gemini 2.0 - Beyond the Chatbot @martonkodok
Results
Conclusions
AI Agents with Gemini 2.0 - Beyond the Chatbot @martonkodok
AI Agents with Gemini 2.0 - Beyond the Chatbot @martonkodok
Enable agentic era
➔ Delegate tasks to agents
➔ Think your ‘dev tasks’ as promptable
➔ Enable them in agentic experiences
➔ New open-source framework from Google designed to simplify the full stack end-to-end
development of agents and multi-agent systems
➔ Use ADK framework to leverage pre-built connectors
➔ Build Multi-Agent Applications
AI Agents with Gemini 2.0 - Beyond the Chatbot @martonkodok
Takeaways
➔ Agents Companion
➔ It extends the original whitepaper by focusing on key
areas needed for production-ready AI agents.
➔ There are insights on multi-agent systems, agent
evaluation techniques, Agent Ops, and Agentic RAG +
more.
➔ Read it here: https://bit.ly/3XIyFpS
AI Agents with Gemini 2.0 - Beyond the Chatbot @martonkodok
Architecting a Multi-Agent System
with Google A2A and ADK
➔ Author: Karl Weinmeister
➔ Youtube channel: Cloud with Karl
➔ Agentic Trading Simulator
➔ Simulator UI (FastAPI) - calls A2A AlphaBot service
➔ AlphaBot (ADK/Python) - receives A2A requests
➔ RiskGuard (ADK/Python) - receives A2A requests
AI Agents with Gemini 2.0 - Beyond the Chatbot @martonkodok
CodeLab - Do it yourself
➔ Google's Agent Stack in Action: ADK, A2A, MCP on Google Cloud
➔ codelabs.developers.google.com/instavibe-adk-multi-agents/instructions
Thank you.
Reea.net - Integrated web solutions driven by creativity
to deliver projects.
#geecon Room 12
Follow for articles:
martonkodok.medium.com
Slides available on:
slideshare.net/martonkodok
Q & A

AI Agents with Gemini 2.0 - Beyond the Chatbot

  • 1.
    AI Agents withGemini 2.0 Beyond the Chatbot Márton Kodok Google Developer Expert at REEA.net #geecon | Room 12
  • 2.
    1. Gemini -The Big Picture 2. Gemini - capabilities 3. Enabling the agentic era 4. Demos 5. Conclusions Agenda AI Agents with Gemini 2.0 - Beyond the Chatbot @martonkodok
  • 3.
    Márton Kodok AI Agentswith Gemini 2.0 - Beyond the Chatbot @martonkodok ● 8x Google Developer Expert on Cloud technologies (2016-->) ● 3x Champion of Google Cloud Innovators program (2021-->) ● Among the Top 3 romanians on Stackoverflow 208k reputation ● Crafting Cloud Architecture+ML backends at REEA.net Articles: martonkodok.medium.com Twitter: @martonkodok Slideshare:martonkodok StackOverflow: pentium10 GitHub: pentium10
  • 4.
    What LLMs are,and are not AI Agents with Gemini 2.0 - Beyond the Chatbot @martonkodok “Large Language Models are not a source of truth.” – Noble Ackerson What they are, however, is a way to transform one thing into another. ● Phrases into code fragments ● Structured data into textual results Combine them with APIs that you already know and work with, and your applications can have amazing new capabilities. A powerful new tool!
  • 5.
    Gemini - TheBig Picture AI Agents with Gemini 2.0 - Beyond the Chatbot @martonkodok Gemini Models ML Other Models Client Libs & SDK Gemini API Vertex AI AI Studio Local Models Gemma ADK AI Tools for Business AI assistant Gemini for Workspace Gemini chat app Gemini Live Notebook LM Developers "Consumers"
  • 6.
    ✅ ✅ ✅ ✅ ✅ @martonkodok Gemini comesin different sizes 2.5 Pro 2.5 Flash 2.0 Pro 1.5 Flash ✅ ❌ Multimodal Input (text, img, video, audio, PDF) Context Window + Output Native Tool Use (Search, Code Exec) Thinking Mode 1M / 8K ❌ ❌ ✅ ✅ ✅ 2M / 8K 1M / 65K 1M / 65K set Thinking Budget 10K free Search queries Preview Preview
  • 7.
    AI Agents withGemini 2.0 - Beyond the Chatbot @martonkodok Texttoimagegeneration withhighqualityandspeed ● Aspectratios 1:1, 9:16, 16:9, 3:4, 4:3 ● Styles photorealism, paintings, richly textured, oil paintings, comic ● UseCases Banners, Posters, Advertizing A lone skateboarder glides through the echoing stillness of an abandoned train station, bathed in the soft glow of a single overhead lamp, creating a dramatic silhouette against the vast, empty space. A dynamic, high-energy animation of a superhero flying across a busy metropolis with a blurred background A watercolor painting of mt fuji A bright style- magazine shot of a woman putting on mascara. Imagen 3
  • 8.
    Demo Gemini: Native imagegeneration AI Agents with Gemini 2.0 - Beyond the Chatbot @martonkodok
  • 9.
    AI Agents withGemini 2.0 - Beyond the Chatbot @martonkodok Veo 2 Veo2
  • 10.
    Vision of abandonedWarsaw, capital of Poland @Tomek Porozynski Veo 2 ● Author: Tomek Porozynski ● GDE & Cloud Architect ● Made with Veo 2
  • 11.
    AI Agents withGemini 2.0 - Beyond the Chatbot @martonkodok ● Local model runs on 1 single GPU ● MultiLingual 140+ languages ● Function Calling ● 1b - 4b - 12b - 27b sizes ● Vision - interleaved image & text input Gemma 3
  • 12.
    Getting started withthe Gemini API AI Agents with Gemini 2.0 - Beyond the Chatbot @martonkodok
  • 13.
    AI Agents withGemini 2.0 - Beyond the Chatbot @martonkodok aistudio.google.com AI Studio ➔ Generate API key ➔ Create, test, and save prompts ➔ SDK in Python/Go/JS+TS/Firebase ➔ Java via LangChain4j library
  • 14.
    AI Agents withGemini 2.0 - Beyond the Chatbot @martonkodok AI Studio ➔ UI is free up to 5-10 RPM ➔ Prompt management ➔ Access to Gemma 3. ➔ Unified SDK references AI Studio mostly as first example
  • 15.
    AI Agents withGemini 2.0 - Beyond the Chatbot @martonkodok console.cloud.google.com/vertex-ai/studio Vertex AI Studio ➔ Service account key ➔ Vertex AI SDK ➔ Access to 120 models via Model Garden ➔ Available for individual/edu/business ➔ Java via google-cloud-vertexai Library
  • 16.
    AI Agents withGemini 2.0 - Beyond the Chatbot @martonkodok console.cloud.google.com/vertex-ai/studio Vertex AI Studio ➔ Project based use ➔ Access to preview models/increased quotas ➔ Help me write prompts ➔ Imagen 3 + Veo 2 ➔ Prompt management
  • 17.
    AI Agents withGemini 2.0 - Beyond the Chatbot @martonkodok gemini.google.com aistudio.google.com ai.google.dev/gemini-api console.cloud.google.com/vertex-ai/studio Consumers Enterprise Developers
  • 18.
    Unified SDK AI Agentswith Gemini 2.0 - Beyond the Chatbot @martonkodok
  • 19.
    Gemini Capabilities AI Agents withGemini 2.0 - Beyond the Chatbot @martonkodok
  • 20.
    What makes Geminiunique? AI Agents with Gemini 2.0 - Beyond the Chatbot @martonkodok Native Tool Use Multimodal in and out Long context length Context Caching Multimodal Live API Thinking Mode Budget driven Spatial Understanding
  • 21.
    AI Agents withGemini 2.0 - Beyond the Chatbot @martonkodok Function Calling API parameters Tools Execution Processing Model Output Function Calls Function Responses Data User Prompt Generating Live Info
  • 22.
    AI Agents withGemini 2.0 - Beyond the Chatbot @martonkodok Google Search Tool google_search_tool = Tool(google_search=GoogleSearch()) response = client.models.generate_content( model="gemini-2.0-flash-001", contents="How’s the weather like today in London?", config=GenerateContentConfig( tools=[google_search_tool] ))
  • 23.
    AI Agents withGemini 2.0 - Beyond the Chatbot @martonkodok Code Execution Tool code_execution_tool = Tool(code_execution=ToolCodeExecution()) response = client.models.generate_content( model="gemini-2.0-flash-001", contents="What is the sum of the first 50 prime numbers?", config=GenerateContentConfig( tools=[code_execution_tool], temperature=0))
  • 24.
    AI Agents withGemini 2.0 - Beyond the Chatbot @martonkodok Spatial Understanding aistudio.google.com/starter-apps/spatial ➔ 2D and 3D spatial understanding ➔ Segmentations masks ➔ Points ➔ 2D and 3D bounding boxes ➔ Local modal alternative: PaliGemma 2
  • 25.
    Gemini 2.0 Enabling theagentic era AI Agents with Gemini 2.0 - Beyond the Chatbot @martonkodok
  • 26.
    AI Agents withGemini 2.0 - Beyond the Chatbot @martonkodok Core Pillars of ADK ➔ Open Source ➔ Code agents like Google ➔ The only dependency is a model ➔ Run on your laptop deploy anywhere ➔ Orchestrate automatically in sequence/parallel ➔ OpenAPI tools ➔ Supports Toolbox, MCPs, A2A Agent Development Kit
  • 27.
    AI Agents withGemini 2.0 - Beyond the Chatbot @martonkodok Build multi-agent capabilities ➔ Function tools ➔ Built-in tools ➔ Third party tools ➔ Agent Garden on Google Cloud ➔ MCP Tools ➔ OpenAPI tools ➔ Agent2Agent Protocol (A2A)
  • 28.
    AI Agents withGemini 2.0 - Beyond the Chatbot @martonkodok ADK vs Langgraph vs LLamaindex ➔ ADK has Gemini and LiteLLM support Langgraph integrates with Langchain ➔ Agent workflow and delegation: ADK uses LLM to decide agent routes Langgraph controlled by stategraph/nodes ➔ Both support short- and long-term memory ADK uses output key to manage state for multiple agents - feels better. ➔ Guardrails: ADK uses callback before/post LLM validation, which can be more centralized control/cleaner Langgraph needs you to manipulate state/LCEL ➔ Llamaindex focused on RAG knowledge assistants Multi-agent documentation from LangGraph.
  • 29.
    AI Agents withGemini 2.0 - Beyond the Chatbot @martonkodok Flexibility in the way you interact with your agents
  • 30.
    Demo 1 ADK agent AIAgents with Gemini 2.0 - Beyond the Chatbot @martonkodok Agent Development Kit
  • 31.
    Gemini 2.0 usecase 🍷wine collection into structured data AI Agents with Gemini 2.0 - Beyond the Chatbot @martonkodok
  • 32.
    AI Agents withGemini 2.0 - Beyond the Chatbot @martonkodok Dataset ➔ Thousands of professional photos ➔ Front + back side of the bottle(s) ➔ High detail photo quality
  • 33.
    AI Agents withGemini 2.0 - Beyond the Chatbot @martonkodok Challenges with Gemini 2.0 ➔ High quality dataset - clear text is visible on zoom ➔ Initial dry run produced ZERO success. What could be the problem? ➔ LLMs works best with rectangle images ➔ Image greater > 384 pixels -> is cropped into tiles... ➔ Each tile is resized to 768x768 ➔ and uses 258 tokens
  • 34.
    AI Agents withGemini 2.0 - Beyond the Chatbot @martonkodok media_resolution config ➔ To control how input images are processed, use the `media_resolution` config parameter: ➔ MEDIA_RESOLUTION_LOW: the image is tokenized into 66 tokens (fixed). ➔ MEDIA_RESOLUTION_MEDIUM: the image is tokenized into 258 tokens (fixed). ➔ MEDIA_RESOLUTION_HIGH: the input image is split (if needed) into non-overlapping crops, leading to X tiles of 258 tokens.
  • 35.
    Gemini Spatial Understanding returningsegmentation masks AI Agents with Gemini 2.0 - Beyond the Chatbot @martonkodok
  • 36.
    AI Agents withGemini 2.0 - Beyond the Chatbot @martonkodok Gemini Spatial Understanding ➔ 2D bounding boxes ➔ Segmentation masks ➔ Points
  • 37.
    AI Agents withGemini 2.0 - Beyond the Chatbot @martonkodok Extract info from images You are an expert in text extraction from images. Your task is to extract all text from the wine bottle label(s) in the **attached images**. Combine the text from both images. Return the text in markdown format, preserving any formatting (e.g., bold, italics, lists) as closely as possible. Use headings (e.g., #, ##) to organize the text if appropriate. Using Gemini 2.0
  • 38.
    AI Agents withGemini 2.0 - Beyond the Chatbot @martonkodok Results
  • 39.
    Conclusions AI Agents withGemini 2.0 - Beyond the Chatbot @martonkodok
  • 40.
    AI Agents withGemini 2.0 - Beyond the Chatbot @martonkodok Enable agentic era ➔ Delegate tasks to agents ➔ Think your ‘dev tasks’ as promptable ➔ Enable them in agentic experiences ➔ New open-source framework from Google designed to simplify the full stack end-to-end development of agents and multi-agent systems ➔ Use ADK framework to leverage pre-built connectors ➔ Build Multi-Agent Applications
  • 41.
    AI Agents withGemini 2.0 - Beyond the Chatbot @martonkodok Takeaways ➔ Agents Companion ➔ It extends the original whitepaper by focusing on key areas needed for production-ready AI agents. ➔ There are insights on multi-agent systems, agent evaluation techniques, Agent Ops, and Agentic RAG + more. ➔ Read it here: https://bit.ly/3XIyFpS
  • 42.
    AI Agents withGemini 2.0 - Beyond the Chatbot @martonkodok Architecting a Multi-Agent System with Google A2A and ADK ➔ Author: Karl Weinmeister ➔ Youtube channel: Cloud with Karl ➔ Agentic Trading Simulator ➔ Simulator UI (FastAPI) - calls A2A AlphaBot service ➔ AlphaBot (ADK/Python) - receives A2A requests ➔ RiskGuard (ADK/Python) - receives A2A requests
  • 43.
    AI Agents withGemini 2.0 - Beyond the Chatbot @martonkodok CodeLab - Do it yourself ➔ Google's Agent Stack in Action: ADK, A2A, MCP on Google Cloud ➔ codelabs.developers.google.com/instavibe-adk-multi-agents/instructions
  • 44.
    Thank you. Reea.net -Integrated web solutions driven by creativity to deliver projects. #geecon Room 12 Follow for articles: martonkodok.medium.com Slides available on: slideshare.net/martonkodok Q & A