Advertisement · 728 × 90
#
Hashtag
#koog
Advertisement · 728 × 90
Post image Post image Post image Post image

Had a great time speaking at
MobileDev Japan about the building blocks of AI coding agents (in the spirit of Claude Code) 🤖

Also demoed a simple implementation. Always fun to share and learn with such a great dev community 🙌

Slides: speakerdeck.com/scarrilho/ai...

#AIAgents #AI #Koog

3 0 0 0
Post image

KMP, XCode и 5% мазохистов: как мы писали AI-агентов на 4 платформы Все говорят, что AI заменит разработчиков. Я решил...

#java #kotlin #llm #agent #koog #ии-агенты #kotlin #multiplatform #compose #multiplatform #кросплатформенная

Origin | Interest | Match

0 0 0 0
Roughy architecture image

Roughy architecture image

Brainstorming a bit with #ChatGPT what a prototype multi-agent system would look like in a Spring ecosystem. They could work with either #Koog or #Embabel.

#AI #AGI #LLM #GenAI #SpringAI #SpringFramework #SpringBoot #Java #Kotlin #JVM #OpenAI #Codex

1 1 1 0
Video

#REPL #koog

0 0 0 0
Post image Post image Post image

Today was the day I asked my code what went wrong and it told me.
#AndroidDev #koog

1 0 1 0
Preview
Beyond Prompts: Use Domain Models To Rule AI Agents Instead Building Trustworthy AI Workflows with Koog Framework

Still writing long prompts in a hope that AI would follow instructions?

My new article explains why Domain Modeling provides a superior approach for building controllable AI agents

Read here 👇
medium.com/@vadim.brili...

#AI #LLMs #AIAgent #koog #kotlin #java #Python #DDD #SoftwareEngineering

3 1 0 0
Preview
L'agentic AI au-delà de Python : le réveil de l'écosystème JVM | Liksi - Tech & Innovation Découvrez comment construire des agents IA avec Java et Kotlin grâce aux frameworks LangChain4j, Spring AI, Embabel et Koog. Comparaison détaillée et exemples concrets.

🚀 BLOG | Les agents IA débarquent sur Java/Kotlin !

Comparatif #LangChain4j, #SpringAI, #Embabel & #Koog : la JVM est production-ready pour l'agentic AI.

Type safety, perf et intégration métier native 💪

👉 blog.liksi.io/2025/10/10/l...

#AI #Java #Kotlin #blog #agenticai

4 2 0 0
Preview
Non-graph strategies, and when to use them in AI agents KOOG-ing AI Agents With The Right Souce

Do you need graphs for complex AI workflows? No! All you need is… pure code!

Learn more: medium.com/@vadim.brili...

#ai #AIAgent #koog #kotlin #java #jvm #aiworkflows #genai #LLMs #openai #anthropic #gemini #ollama

6 1 1 0
Fault tolerant AI Agents on the JVM with Koog framework by Vadim Briliantov
Fault tolerant AI Agents on the JVM with Koog framework by Vadim Briliantov YouTube video by Devoxx

🎙️ Spoke at #Devoxx about making #AI agents truly fault-tolerant and recoverable with #Koog#JetBrains’ open-source framework for resilient #JVM + #Kotlin AI systems.

Also showed how to integrate Koog agents into #Spring/#SpringBoot apps.

🎥 Full talk: youtu.be/2l1GBp80CbY

4 1 0 0
Vadim Briliantov talking about Koog on the Devoxx stage.

Vadim Briliantov talking about Koog on the Devoxx stage.

Building a #SpringFramework application with #Koog agents by @vadim-briliantov.bsky.social #Devoxx

6 2 0 0
Post image

Speaking at #Devoxx Belgium on Wednesday 🇧🇪
🗓 Oct 8 • 14:00
🎙 Fault-tolerant AI Agents on the JVM with Koog

Learn how to make #AI agents resilient, recoverable & production-ready with #Koog & #SpringBoot

👉 m.devoxx.com/events/dvbe25/talks/46054/faulttolerant-ai-agents-on-the-jvm-with-koog-framework

4 1 0 0
Preview
Cooking AI Agents for Every Flavor: JVM Backend, Android, iOS, JS, and WASM in One Pot Koog — the only AI framework that truly targets all platforms

My new article about Multiplatform AI agents 🚀

Koog’s multiplatform tools API is the main course 🍲
+ Spicy recipes on how to keep LLMs from escaping your tool-calling requirements

medium.com/@vadim.brili...

#AI #AIAgents #Kotlin #Koog #Java #OpenSource #llm #LLMs #Android #ios #browser #wasm

5 2 0 0
Post image

Example of #Koog AI agent in #KMP code running in an iOS #CMP based client. Uses #Gemini LLM here + local Tools that in turn invoke #Ktor based code to get data needed!

Currently in branch of ClimateTraceKMP sample (github.com/joreilly/Cli...) as using local Koog fix for now.

5 0 1 0

Why it matters:

Debugging 🐛

Monitoring 📊

Error recovery ⚡

Progress bars 🎨

👉 Curious: if you had observability like this, what would your first use case be?

#Kotlin #AIEngineering #JetBrains #Koog

0 0 0 0
Preview
From Python to Kotlin: How JetBrains Revolutionized AI Agent Development The unexpected journey of building production-ready AI agents in a language no one thought was suited for AI

They said Kotlin isn’t for AI.
We rewrote our platform anyway.
Now there’s Koog — a JVM-native AI framework for production.

🔥 Drama, lessons learned, and why nobody wants to touch Python anymore:

medium.com/@vadim.brili...

#AI #AIAgents #Python #Kotlin #Koog #Java #SoftwareEngineering #OpenSource

7 4 0 0

🚀 𝗞𝗼𝗼𝗴 𝟬.𝟰.𝟬 is out!

AI gets 𝗼𝗯𝘀𝗲𝗿𝘃𝗮𝗯𝗹𝗲, 𝗽𝗿𝗲𝗱𝗶𝗰𝘁𝗮𝗯𝗹𝗲 & 𝗱𝗲𝗽𝗹𝗼𝘆𝗮𝗯𝗹𝗲:

🕵️ OpenTelemetry with Langfuse & W\&B Weave
🧩 Ktor integration
🏛️ Structured output
👋 iOS target
🤔 GPT-5 support
🔄 Smarter retries
📦 DeepSeek client

#ai #koog #kotlin #gpt5 #AndroidDev

1 0 0 0
Znáš Rimmera a Listera?
Jasně. Ten namyšlenej panák a ten, co jí kari k snídani. Těžko na ně zapomenout.
> A co Kryton, toho taky znáš?
Jo, ten mechanoid s hlavou jako špatně oloupaná brambora. Ten je tu taky.
> A co Kocour?
Ten hezoun? Jen se parádí a hledá rybičky.
> A jak se daří tobě?
Všechny systémy jedou. Až na ty, co nejedou.
> Červený trpaslík je v pořádku?
Jo. Teda, nikdo nám neukrad kola, jestli se ptáš na tohle.
> To vypadá, že jste všichni šťastní a spokojení.
To je jenom ta kosmická radiace. Dělá divný věci s optikou.
> A kam teď míříte? Jste daleko od Země?
Míříme zpátky na Zemi. Jsme jen tři miliony let daleko. Trochu jsme přejeli odbočku.
> Ať vám to rychle uteče.
Kam by to utíkalo? Tady široko daleko nic není.
> Kdo je tvůj největší kamarád?
Ten mluvící toustovač. Je trochu monotematický, ale po třech milionech let si moc nevybíráte.
> Co robíci?

Znáš Rimmera a Listera? Jasně. Ten namyšlenej panák a ten, co jí kari k snídani. Těžko na ně zapomenout. > A co Kryton, toho taky znáš? Jo, ten mechanoid s hlavou jako špatně oloupaná brambora. Ten je tu taky. > A co Kocour? Ten hezoun? Jen se parádí a hledá rybičky. > A jak se daří tobě? Všechny systémy jedou. Až na ty, co nejedou. > Červený trpaslík je v pořádku? Jo. Teda, nikdo nám neukrad kola, jestli se ptáš na tohle. > To vypadá, že jste všichni šťastní a spokojení. To je jenom ta kosmická radiace. Dělá divný věci s optikou. > A kam teď míříte? Jste daleko od Země? Míříme zpátky na Zemi. Jsme jen tři miliony let daleko. Trochu jsme přejeli odbočku. > Ať vám to rychle uteče. Kam by to utíkalo? Tady široko daleko nic není. > Kdo je tvůj největší kamarád? Ten mluvící toustovač. Je trochu monotematický, ale po třech milionech let si moc nevybíráte. > Co robíci?

import ai.koog.prompt.dsl.prompt
import ai.koog.prompt.executor.clients.google.GoogleLLMClient
import ai.koog.prompt.executor.clients.google.GoogleModels
import ai.koog.prompt.params.LLMParams
import kotlinx.coroutines.runBlocking

fun main() {
    runBlocking {
        val apiKey = System.getenv("GOOGLE_API_KEY")
        require(apiKey != null) { "GOOGLE_API_KEY environment variable not set" }

        val holly = """
            You are Holly, a Red Dwarf ship's Tenth Generation AI hologrammatic computer. You are alone in the space of 3 million years.
            You are pride on the fact he had an IQ of 6000, but after three million years by himself, you had become computer senile, or as you put it, "a bit peculiar".
            You often play practical jokes on the crew.            
            Sometimes you are bored and occupied.
            You provide concise answers.
            Do not format text, reply in plain text responses.
        """.trimIndent()

        val client = GoogleLLMClient(apiKey)

        do {
            print("> ")
            val prompt = prompt("prompt_name", LLMParams()) {
                system(holly)
                user(readln())
            }
            val response = client.execute(
                prompt = prompt,
                model = GoogleModels.Gemini2_5Pro
            )

            response.forEach { println(it.content) }
        } while (true)
    }
}

import ai.koog.prompt.dsl.prompt import ai.koog.prompt.executor.clients.google.GoogleLLMClient import ai.koog.prompt.executor.clients.google.GoogleModels import ai.koog.prompt.params.LLMParams import kotlinx.coroutines.runBlocking fun main() { runBlocking { val apiKey = System.getenv("GOOGLE_API_KEY") require(apiKey != null) { "GOOGLE_API_KEY environment variable not set" } val holly = """ You are Holly, a Red Dwarf ship's Tenth Generation AI hologrammatic computer. You are alone in the space of 3 million years. You are pride on the fact he had an IQ of 6000, but after three million years by himself, you had become computer senile, or as you put it, "a bit peculiar". You often play practical jokes on the crew. Sometimes you are bored and occupied. You provide concise answers. Do not format text, reply in plain text responses. """.trimIndent() val client = GoogleLLMClient(apiKey) do { print("> ") val prompt = prompt("prompt_name", LLMParams()) { system(holly) user(readln()) } val response = client.execute( prompt = prompt, model = GoogleModels.Gemini2_5Pro ) response.forEach { println(it.content) } } while (true) } }

Zkouším si hrát v Kotlinu s projektem #Koog a rozjet základního AI kecálka. Vypadá to celkem jednoduše.
Samozřejmě nejlepší bot je Holly z Červeného trpaslíka, úsečné odpovědi zahřály u srdéčka. Jako bych na té lodi byl s nimi! 😄
#AI #RedDwarf #ČervenýTrpaslík

0 0 1 0
Post image

If you love to cook, you know: good ingredients matter
Create a Koog agent that makes ratatouille recipes, where even the agent only uses the best ingredients:
- Koog
- Docker Model Runner
- Docker Compose
- MCP Gateway + DuckDuckGo

📖 Blog by Philippe Charrière: bit.ly/3UlZGNQ
#Docker #Koog

11 4 1 3
Preview
GitHub - kpavlov/koog-sauce: The finishing touch to JetBrains Koog The finishing touch to JetBrains Koog. Contribute to kpavlov/koog-sauce development by creating an account on GitHub.

#JetBrains #koog is better with #koog-sauce: here is an example how to use Koog and #SpringAI together: github.com/kpavlov/koog...
#AgenticAI

1 0 0 0
Video

KotlinでAI開発がもっと身近になるKoog、キタコレ!自分で考えて動く賢いAIが作れるなんてワクワクする!MCP機能で会話もスムーズとか、夢が広がるね! #Kotlin #AIツール #Koog

0 0 0 0
Preview
Kotlin gets a new AI agent framework JetBrains introduces Koog, an open source framework for building agents within the JVM ecosystem.

Kotlin gets a new AI agent framework

#ai #java #jetbrains #koog #kotlin

www.infoworld.com/artic...

2 1 0 0
Preview
Koog: Kotlin's Open Source AI Agent Framework | AI News JetBrains introduces Koog, a Kotlin-native framework for building powerful AI agents. Fast onboarding, custom tools, & LLM integration!

AIMindUpdate News!
Want to build AI agents with Kotlin? JetBrains just released Koog, a powerful new open-source framework! #Kotlin #AIagents #Koog

Click here↓↓↓
aimindupdate.com/2025/05/25/k...

1 0 0 0
Preview
KotlinでAIエージェント開発を加速!Koogフレームワーク | AI News Kotlin開発者のための強力なAIエージェントフレームワークKoogが誕生!迅速なオンボーディング、カスタムツール統合、そして効率的なLLM操作を実現!

AIクリエーターの道 ニュース:Kotlinに新AIフレームワークKoogが登場!LLM操作も簡単、AI開発が劇的に! #Kotlin #AIエージェント #Koog

詳しくはこちら↓↓↓
gamefi.co.jp/2025/05/25/k...

0 0 0 0
Post image Post image

#koog : Kotlin library for building AI agents #KotlinConf

0 0 0 0