Ollama x FriendliAI Seoul Meetup์˜ ์ปค๋ฒ„ ์ด๋ฏธ์ง€
Ollama x FriendliAI Seoul Meetup์˜ ์ปค๋ฒ„ ์ด๋ฏธ์ง€
Avatar for FriendliAI
์ฃผ์ตœ
FriendliAI
์ฃผ์ตœ์ž
๋น„๊ณต๊ฐœ ์ด๋ฒคํŠธ

Ollama x FriendliAI Seoul Meetup

์ฃผ์†Œ ํ™•์ธ์„ ์œ„ํ•ด ๋“ฑ๋กํ•˜๊ธฐ
๋Œ€ํ•œ๋ฏผ๊ตญ์„œ์šธํŠน๋ณ„์‹œ
๋“ฑ๋ก
์Šน์ธ ํ•„์š”
๋“ฑ๋ก์€ ํ˜ธ์ŠคํŠธ์˜ ์Šน์ธ์ด ํ•„์š”ํ•ฉ๋‹ˆ๋‹ค.
ํ™˜์˜ํ•ฉ๋‹ˆ๋‹ค! ์ด๋ฒคํŠธ์— ์ฐธ์—ฌํ•˜๋ ค๋ฉด ์•„๋ž˜์—์„œ ๋“ฑ๋กํ•ด ์ฃผ์„ธ์š”.
์ด๋ฒคํŠธ ์†Œ๊ฐœ

โ€‹Ollama x FriendliAI Seoul Meetup ๐Ÿ‡ฐ๐Ÿ‡ท
: Building Production Agents with Open Models

โ€‹GLM, Kimi, DeepSeek, MiniMax, Nemotron ๋“ฑ ํ”„๋ก ํ‹ฐ์–ด ์˜คํ”ˆ ๋ชจ๋ธ์ด ๋น ๋ฅด๊ฒŒ ๋ฐœ์ „ํ•˜๊ณ  ์žˆ์Šต๋‹ˆ๋‹ค. ์ด์ œ ๊ฐœ๋ฐœ์ž์—๊ฒŒ ์ค‘์š”ํ•œ ์งˆ๋ฌธ์€ "์–ด๋–ค ๋ชจ๋ธ์„ ์“ธ ๊ฒƒ์ธ๊ฐ€"๋ฅผ ๋„˜์–ด, "๊ทธ ๋ชจ๋ธ์„ ํ”„๋กœ๋•์…˜์ด ์š”๊ตฌํ•˜๋Š” ์†๋„ยท์•ˆ์ •์„ฑยท๋น„์šฉ ํšจ์œจ๋กœ AI ์—์ด์ „ํŠธ์— ์–ด๋–ป๊ฒŒ ์šด์˜ํ•  ๊ฒƒ์ธ๊ฐ€"์ž…๋‹ˆ๋‹ค.

โ€‹์›ํ•˜๋Š” ์˜คํ”ˆ ๋ชจ๋ธ์„ ์„ ํƒํ•˜๊ณ , AI ์—์ด์ „ํŠธ๋ฅผ ๋งŒ๋“ค๊ณ , ํ”„๋กœ๋•์…˜๊นŒ์ง€ ๊ฐ€์ ธ๊ฐ€์„ธ์š”. Ollama์™€ FriendliAI๊ฐ€ ์˜คํ”ˆ ๋ชจ๋ธ๋กœ AI ์—์ด์ „ํŠธ๋ฅผ ๊ตฌ์ถ•ํ•˜๊ณ  ์—์ด์ „ํ‹ฑ ์›Œํฌ๋กœ๋“œ๋ฅผ ํ”„๋กœ๋•์…˜ ๊ทœ๋ชจ๋กœ ํ™•์žฅํ•˜๋Š” ๊ธฐ์ˆ ๊ณผ ์‹ค์ œ ์‚ฌ๋ก€๋ฅผ ๊ณต์œ ํ•ฉ๋‹ˆ๋‹ค. ๋ฐ๋ชจ๋ฅผ ๋„˜์–ด ์‹ค์ œ ์„œ๋น„์Šค์—์„œ ์˜คํ”ˆ ๋ชจ๋ธ์„ ์šด์˜ํ•ด ๋ณธ ๋นŒ๋”๋“ค๊ณผ ๋งŒ๋‚˜ ๊ฒฝํ—˜๊ณผ ์ธ์‚ฌ์ดํŠธ๋ฅผ ๋‚˜๋ˆ„์„ธ์š”.

โ€‹(๋ฐœํ‘œ ์ฃผ์ œ์™€ ์—ฐ์‚ฌ๋Š” ์ถ”ํ›„ ์—…๋ฐ์ดํŠธ ์˜ˆ์ •์ž…๋‹ˆ๋‹ค.)

โ€‹๋‹ค๋ฃฐ ๋‚ด์šฉ

  • โ€‹์˜คํ”ˆ ๋ชจ๋ธ๋กœ AI ์—์ด์ „ํŠธ๋ฅผ ์„ค๊ณ„ยท๊ตฌ์ถ•ํ•˜๋Š” ์‹ค์ œ ์›Œํฌํ”Œ๋กœ์šฐ

  • โ€‹๋กœ์ปฌ ๊ฐœ๋ฐœ ํ™˜๊ฒฝ์—์„œ ํ”„๋กœ๋•์…˜ ์ถ”๋ก ๊นŒ์ง€, ์—์ด์ „ํŠธ๋ฅผ ๋ฐฐํฌํ•˜๋Š” ๋ฐฉ๋ฒ•

  • โ€‹์—์ด์ „ํ‹ฑ ์›Œํฌ๋กœ๋“œ์˜ ์ง€์—ฐ ์‹œ๊ฐ„ยท์ฒ˜๋ฆฌ๋Ÿ‰ยท๋น„์šฉ์„ ์ตœ์ ํ™”ํ•˜๋Š” ์‹ค์ „ ์‚ฌ๋ก€

  • โ€‹์˜คํ”ˆ ๋ชจ๋ธ ์„ ํƒ ๊ธฐ์ค€๊ณผ ํ”„๋กœ๋•์…˜ ์šด์˜์—์„œ์˜ ํŠธ๋ ˆ์ด๋“œ์˜คํ”„

  • โ€‹์ฐธ์„์ž Q&A ๋ฐ ๋นŒ๋” ๋„คํŠธ์›Œํ‚น

โ€‹ํ”„๋กœ๊ทธ๋žจ

  • โ€‹18:00 ~ 19:00, ์ฒดํฌ์ธ ๋ฐ ๋„คํŠธ์›Œํ‚น

  • โ€‹19:00 ~ 19:10, ํ™˜์˜ ์ธ์‚ฌ ๋ฐ ์˜คํ”„๋‹ (Ollama ยท FriendliAI)

  • โ€‹19:10 ~ 20:10, ๋ฐœํ‘œ ์„ธ์…˜

    • โ€‹Jeffrey Morgan, CEO of Ollama

    • โ€‹Byung-Gon Chun, CEO of FriendliAI

    • โ€‹TBD (๊ฒŒ์ŠคํŠธ ์„ธ์…˜)

  • โ€‹20:10 ~ 20:30, ์˜คํ”ˆ Q&A

  • โ€‹20:30 ~ 21:00, ๋„คํŠธ์›Œํ‚น ๋ฐ ๋งˆ๋ฌด๋ฆฌ

โ€‹(์„ธ๋ถ€ ์ผ์ •์€ ๋ณ€๊ฒฝ๋  ์ˆ˜ ์žˆ์Šต๋‹ˆ๋‹ค)

โ€‹์ด๋Ÿฐ ๋ถ„๋“ค์„ ๊ธฐ๋‹ค๋ฆฝ๋‹ˆ๋‹ค

  • โ€‹์˜คํ”ˆ ๋ชจ๋ธ๋กœ AI ์—์ด์ „ํŠธ๋‚˜ LLM ์• ํ”Œ๋ฆฌ์ผ€์ด์…˜์„ ๋งŒ๋“ค๊ณ  ์žˆ๊ฑฐ๋‚˜ ๋งŒ๋“ค ๊ณ„ํš์ธ ๋ถ„

  • โ€‹AI/ML ์—”์ง€๋‹ˆ์–ด, ๋ฐฑ์—”๋“œยทํ”Œ๋žซํผ ์—”์ง€๋‹ˆ์–ด, ์—”์ง€๋‹ˆ์–ด๋ง ๋ฆฌ๋”

  • โ€‹AI ์ œํ’ˆ์„ ๋งŒ๋“œ๋Š” ํ…Œํฌ ์ฐฝ์—…์ž ๋ฐ ์ดˆ๊ธฐ ํŒ€

  • โ€‹์˜คํ”ˆ ๋ชจ๋ธ์„ ํ”„๋กœ๋•์…˜์— ์˜ฌ๋ฆฌ๋ฉฐ ๊ฒช์€ ์‹œํ–‰์ฐฉ์˜ค๋ฅผ ๋‚˜๋ˆ„๊ณ  ์‹ถ์€ ๋ถ„

โ€‹Ollama๋‚˜ FriendliAI๋ฅผ ์จ๋ณธ ๊ฒฝํ—˜์€ ํ•„์ˆ˜๊ฐ€ ์•„๋‹™๋‹ˆ๋‹ค. ์ง€๊ธˆ ๋งŒ๋“ค๊ณ  ์žˆ๋Š” ๊ฒƒ, ํ’€๊ณ  ์‹ถ์€ ๋ฌธ์ œ๋ฅผ ํ•จ๊ป˜ ๊ฐ€์ ธ์™€ ์ฃผ์„ธ์š”.

โ€‹์ฐธ๊ฐ€ ์•ˆ๋‚ด

  • โ€‹์ฐธ๊ฐ€๋น„: ๋ฌด๋ฃŒ

  • โ€‹์ธ์› ์ œํ•œ์ด ์žˆ์–ด ๋“ฑ๋ก ํ›„ ์Šน์ธ๋œ ๋ถ„๊ป˜ ์ฐธ๊ฐ€ ํ™•์ • ์•ˆ๋‚ด๋ฅผ ๋“œ๋ฆฝ๋‹ˆ๋‹ค.

  • โ€‹๋ฐœํ‘œ ์–ธ์–ด: ํ•œ๊ตญ์–ดยท์˜์–ด

  • โ€‹๊ฐ„๋‹จํ•œ ์Œ๋ฃŒ์™€ ๋‹ค๊ณผ๊ฐ€ ์ œ๊ณต๋ฉ๋‹ˆ๋‹ค.

โ€‹์žฅ์†Œ


โ€‹ํ˜ธ์ŠคํŠธ ์†Œ๊ฐœ

โ€‹Ollama

โ€‹Ollama์™€ ํ•จ๊ป˜๋ผ๋ฉด ์ตœ์‹  ์˜คํ”ˆ ๋ชจ๋ธ์„ ์›ํ•˜๋Š” ํ™˜๊ฒฝ์—์„œ ์†์‰ฝ๊ฒŒ ์‹คํ–‰ํ•  ์ˆ˜ ์žˆ์Šต๋‹ˆ๋‹ค.

โ€‹๋ช‡ ๋ถ„ ๋งŒ์— ๊ฐœ๋ฐœ์„ ์‹œ์ž‘ํ•˜๊ณ , ๋ฐ์ดํ„ฐ๋ฅผ ์•ˆ์ „ํ•˜๊ฒŒ ๋ณดํ˜ธํ•˜๋ฉฐ, ์ž‘์€ ์‹คํ—˜์—์„œ ์‹ค์ œ ํŠธ๋ž˜ํ”ฝ์„ ์ฒ˜๋ฆฌํ•˜๋Š” ์„œ๋น„์Šค๊นŒ์ง€ ํ™•์žฅํ•  ์ˆ˜ ์žˆ์Šต๋‹ˆ๋‹ค. ์ˆ˜๋ฐฑ๋งŒ ๋ช…์˜ ๊ฐœ๋ฐœ์ž๋“ค์ด ์˜คํ”ˆ ๋ชจ๋ธ์„ ์‹œ์ž‘ํ•˜๋Š” ๊ฐ€์žฅ ์‰ฌ์šด ๋ฐฉ๋ฒ•์œผ๋กœ Ollama๋ฅผ ์‚ฌ์šฉํ•˜๊ณ  ์žˆ์Šต๋‹ˆ๋‹ค.

โ€‹FriendliAI

โ€‹FriendliAI๋Š” ์˜คํ”ˆ ๋ชจ๋ธ์„ ํ”„๋กœ๋•์…˜์—์„œ ์„œ๋น™ํ•˜๋Š” ํ”„๋ก ํ‹ฐ์–ด AI ์ถ”๋ก  ํด๋ผ์šฐ๋“œ์ž…๋‹ˆ๋‹ค. Model APIs, Dedicated Endpoints, BYOG(Bring Your Own GPU)๋ฅผ ํ†ตํ•ด ๊ฐœ๋ฐœ์ž์™€ ๊ธฐ์—…์ด ์›ํ•˜๋Š” ์˜คํ”ˆ ๋ชจ๋ธ์„ ๋น ๋ฅด๊ณ  ์•ˆ์ •์ ์ด๋ฉฐ ๋น„์šฉ ํšจ์œจ์ ์œผ๋กœ ์šด์˜ํ•  ์ˆ˜ ์žˆ๋„๋ก ์ง€์›ํ•ฉ๋‹ˆ๋‹ค.


โ€‹Ollama x FriendliAI Seoul Meetup ๐Ÿ‡ฐ๐Ÿ‡ท
: Building Production Agents with Open Models

โ€‹Frontier open models like GLM, Kimi, DeepSeek, MiniMax, and Nemotron are advancing rapidly. For developers, the question is no longer just which model to use, but how to run that model for AI agents with the speed, reliability, and cost efficiency required in production.

โ€‹Choose the open model you want. Build your AI agent. Take it to production. Join Ollama and FriendliAI to explore the technologies and real-world practices behind building AI agents with open models and scaling agentic workloads for production. Connect with builders who have taken open models beyond the demo and into real services, and share experiences and insights.

โ€‹(Talks and speakers will be announced soon.)

โ€‹What we'll cover

  • โ€‹Real-world workflows for designing and building AI agents on open models

  • โ€‹Taking agents from local development to production inference

  • โ€‹Practical lessons on optimizing latency, throughput, and cost for agentic workloads

  • โ€‹How to choose an open model, and the trade-offs that show up in production

  • โ€‹Open Q&A and networking with fellow builders

โ€‹Program

  • โ€‹18:00 โ€“ 19:00 PM, Check-in & networking

  • โ€‹19:00 โ€“ 19:10 PM, Welcome & opening remarks (Ollama ยท FriendliAI)

  • โ€‹19:10 โ€“ 20:10 PM, Talks

    • โ€‹Jeffrey Morgan, CEO of Ollama

    • โ€‹Byung-Gon Chun, CEO of FriendliAI

    • โ€‹TBD (Guest session)

  • โ€‹20:10 โ€“ 20:30 PM, Open Q&A

  • โ€‹20:30 โ€“ 21:00 PM, Networking & wrap-up

โ€‹(Schedule subject to change)

โ€‹Who should attend

  • โ€‹Anyone building โ€” or planning to build โ€” AI agents or LLM applications on open models

  • โ€‹AI/ML engineers, backend and platform engineers, engineering leaders

  • โ€‹Technical founders and early teams building AI products

  • โ€‹Anyone who wants to share the hard-won lessons of running open models in production

โ€‹No prior experience with Ollama or FriendliAI is required. Bring what you're building and the problems you're trying to solve.

โ€‹Registration

  • โ€‹Admission: Free

  • โ€‹Seats are limited. Registrations are subject to approval; confirmed attendees will be notified.

  • โ€‹Session language: Koreanยท English

  • โ€‹Light refreshments will be provided.

โ€‹Venue


โ€‹About the hosts

โ€‹Ollama

โ€‹Ollama lets you run today's leading open models wherever you need them.

Start building in minutes, keep your data secure and grow from a first experiment to applications serving real traffic. Millions of developers have made Ollama their starting point for building with open models.

โ€‹FriendliAI

โ€‹FriendliAI is a frontier AI inference cloud serving open-weight models in production. Through Model APIs, Dedicated Endpoints, and BYOG (Bring Your Own GPU), FriendliAI helps developers and enterprises run the open models they choose with the speed, reliability, and cost efficiency production demands.

์žฅ์†Œ
์ด๋ฒคํŠธ์˜ ์ •ํ™•ํ•œ ์œ„์น˜๋ฅผ ํ™•์ธํ•˜๋ ค๋ฉด ๋“ฑ๋กํ•ด ์ฃผ์„ธ์š”.
๋Œ€ํ•œ๋ฏผ๊ตญ์„œ์šธํŠน๋ณ„์‹œ
Avatar for FriendliAI
์ฃผ์ตœ
FriendliAI
์ฃผ์ตœ์ž