Loading video...

Video Failed to Load

Go Home

Made a UI to visualize multiple AI agents playing as different civilizations, each navigating its own decision tree Most agent workflow UIs are pretty static, so I wanted to build something that feels more dynamic and fun to explore

124,083 views • 3 months ago •via X (Twitter)

47 Comments

XIN Chen's profile picture
XIN Chen3 months ago

For Drug Design

Mikhail Kashkin's profile picture
Mikhail Kashkin3 months ago

UI is sleek and beautiful, love it!

Dilum Sanjaya's profile picture
Dilum Sanjaya3 months ago

thanks Mikhail

Nick Venturi's profile picture
Nick Venturi3 months ago

basically a digital ant farm 🤤

mars's profile picture
mars3 months ago

this is so cool!

Dilum Sanjaya's profile picture
Dilum Sanjaya3 months ago

thanks

Matthieu Talbot's profile picture
Matthieu Talbot3 months ago

Il me faut ça, pour ma vie; je parle à mon agent le soir, et à la fin de la semaine il me fait un récap en mode Civilisation Tech Tree.

gil's profile picture
gil3 months ago

Like others have said, this is very well thought out and makes me realize I’ve been lazy in my UX approach to drawing graphs. I don’t have to just take vue flow ootb. Well done.

Dilum Sanjaya's profile picture
Dilum Sanjaya3 months ago

thanks!, That's why I choose topics like these for my demos. When working with real data, it gets boring pretty quickly, so there are definitely times when I also just use whatever's available ootb and call it a day

Sofiia Matsiutsia's profile picture
Sofiia Matsiutsia3 months ago

this is genius 🤌

Dilum Sanjaya's profile picture
Dilum Sanjaya3 months ago

thanks Sofiia!

niccomi 🧙‍♂️'s profile picture
niccomi 🧙‍♂️3 months ago

civ gamers will rejoice

Dilum Sanjaya's profile picture
Dilum Sanjaya3 months ago

it is heavily inspired by the civ tech tree🙂

KS's profile picture
KS3 months ago

What was your approach for this? I'm doing something similar for my custom dashboard/mission control but I've found throughput tanks as soon as I ask for custom graphical assets, textures etc.

Dillon's profile picture
Dillon3 months ago

Age of Agents

ChronoAI's profile picture
ChronoAI3 months ago

This is the same idea we built consensus-loop around — different agents playing different roles, sealed from each other, meta-judge converges. Got it running 24/7 on real repo issues now: solvers + reviewers actually ship PRs and releases on their own.

SecWay's profile picture
SecWay3 months ago

有意思,值得欣赏

Dilum Sanjaya's profile picture
Dilum Sanjaya3 months ago

谢谢

hlitoroy (opinion arc)'s profile picture
hlitoroy (opinion arc)3 months ago

another ai slop

Saaga's profile picture
Saaga3 months ago

Really cool UI

🤖 🇺🇸 ²'s profile picture
🤖 🇺🇸 ²3 months ago

Love this! Any chance you’d release it for others to try?

Crazyox🌶️(微微辣版)'s profile picture
Crazyox🌶️(微微辣版)3 months ago

这个好玩,如果你能让硅基文明和碳基文明交汇并且自然发展出多种结局,我真的会想看。

Quinn Favo's profile picture
Quinn Favo3 months ago

Pretty cool!

Dilum Sanjaya's profile picture
Dilum Sanjaya3 months ago

thanks

Brandon Bartlett's profile picture
Brandon Bartlett3 months ago

This is pretty cool. I had a similar idea. Making it like what happens if this or that choice is made then what the downstream effects are

Trog's profile picture
Trog3 months ago

Thats pretty cool. Want to teach a man to fish?

Matt Nicosia's profile picture
Matt Nicosia3 months ago

This is awesome

Dilum Sanjaya's profile picture
Dilum Sanjaya3 months ago

thanks Matt

0XNEET's profile picture
0XNEET3 months ago

Really love your content. It's so inspiring

Dilum Sanjaya's profile picture
Dilum Sanjaya3 months ago

thanks

Leo Ge's profile picture
Leo Ge3 months ago

Cool!!!

Dilum Sanjaya's profile picture
Dilum Sanjaya3 months ago

thanks Leo

leolee's profile picture
leolee3 months ago

This is the kind of UI that makes agent systems feel less like dashboards and more like a world you can actually read. The civilization framing is smart too — it gives each agent a clear identity instead of a pile of nodes doing random stuff. Static workflow UIs get old fast. This feels easier to explore, and probably easier to debug when things go sideways. Nice balance of useful and fun 🔥

Tyler Mayberry's profile picture
Tyler Mayberry3 months ago

This is really cool making me come up with a lot of new ideas.

Dilum Sanjaya's profile picture
Dilum Sanjaya3 months ago

thanks Tyler!

Yann André's profile picture
Yann André3 months ago

I love that. 👍🏻

Dilum Sanjaya's profile picture
Dilum Sanjaya3 months ago

thanks Yann

Thành's profile picture
Thành2 months ago

This is the coolest agent

AI Mastery Guide's profile picture
AI Mastery Guide3 months ago

Visualizing agent decision trees as a live civilization map makes multi-agent workflows actually fun to watch unfold 🗺️

Pan's's profile picture
Pan's3 months ago

really cool

Bruno Serrate → postyourmind.com's profile picture
Bruno Serrate → postyourmind.com3 months ago

Pretty cool UI. This could be easy a UI game or something lol

Dilum Sanjaya's profile picture
Dilum Sanjaya3 months ago

thanks, it is inspired by Civilization games

Samian's profile picture
Samian3 months ago

visualizing the decision tree branching in real time would be the win. like actually seeing each agent weigh its options before committing, not just the final move

Dr. Xi Zeng's profile picture
Dr. Xi Zeng3 months ago

The agent map is the right instinct. Most workflow screens make agents look like boxes waiting in line. A game-like view can show the part users need: which agent saw what, chose what, and got stuck where.

paleocybernetic's profile picture
paleocybernetic3 months ago

oh my gosh please open source so I can update my AI advisor council with this

Gene's profile picture
Gene3 months ago

Mmm interesting

Harley Lewis Foote's profile picture
Harley Lewis Foote3 months ago

Looks fun. Most agent UIs still feel like a spreadsheet with extra steps.

Related Videos

New Course: ACP: Agent Communication Protocol Learn to build agents that communicate and collaborate across different frameworks using ACP in this short course built with IBM Research's BeeAI, and taught by Sandi Besen, AI Research Engineer & Ecosystem Lead at IBM, and Nicholas Renotte, Head of AI Developer Advocacy at IBM. Building a multi-agent system with agents built or used by different teams and organizations can become challenging. You may need to write custom integrations each time a team updates their agent design or changes their choice of agentic orchestration framework. The Agent Communication Protocol (ACP) is an open protocol that addresses this challenge by standardizing how agents communicate, using a unified RESTful interface that works across frameworks. In this protocol, you host an agent inside an ACP server, which handles requests from an ACP client and passes them to the appropriate agent. Using a standardized client-server interface allows multiple teams to reuse agents across projects. It also makes it easier to switch between frameworks, replace an agent with a new version, or update a multi-agent system without refactoring the entire system. In this course, you’ll learn to connect agents through ACP. You’ll understand the lifecycle of an ACP Agent and how it compares to other protocols, such as MCP (Model Context Protocol) and A2A (Agent-to-Agent). You’ll build ACP-compliant agents and implement both sequential and hierarchical workflows of multiple agents collaborating using ACP. Through hands-on exercises, you’ll build: - A RAG agent with CrewAI and wrap it inside an ACP server. - An ACP Client to make calls to the ACP server you created. - A sequential workflow that chains an ACP server, created with Smolagents, to the RAG agent. - A hierarchical workflow using a router agent that transforms user queries into tasks, delegated to agents available through ACP servers. - An agent that uses MCP to access tools and ACP to communicate with other agents. You’ll finish up by importing your ACP agents into the BeeAI platform, an open-source registry for discovering and sharing agents. ACP enables collaboration between agents across teams and organizations. By the end of this course, you’ll be able to build ACP agents and workflows that communicate and collaborate regardless of framework. Please sign up here:

Andrew Ng

105,758 views • 1 year ago