Machine Learning Engineer (Agentic AI)

furiosa-aiยท Software
Apply Now โ†—
๐Ÿ“ Seoul HQFullTime

About this role

About AI Transformation Team

AI Transformation Team์€ AI๋ฅผ ๊ธฐ๋ฐ˜์œผ๋กœ ์ƒˆ๋กœ์šด ์—”์ง€๋‹ˆ์–ด๋ง ๋ฌธ์ œ ํ•ด๊ฒฐ ๋ฐฉ์‹๊ณผ workflow๋ฅผ ๋งŒ๋“ค์–ด๊ฐ€๋Š” ํŒ€์ž…๋‹ˆ๋‹ค. ๋‹ค์–‘ํ•œ AI ๊ธฐ์ˆ ์„ ํ™œ์šฉํ•ด ์—”์ง€๋‹ˆ์–ด๋ง ๋ฌธ์ œ๋ฅผ ์ž์œจ์ ์œผ๋กœ ํƒ์ƒ‰ํ•˜๊ณ  ๊ฐœ์„ ํ•  ์ˆ˜ ์žˆ๋Š” ์‹œ์Šคํ…œ์„ ๊ฐœ๋ฐœํ•˜๊ณ  ์žˆ์œผ๋ฉฐ, ์†Œํ”„ํŠธ์›จ์–ด, ํ•˜๋“œ์›จ์–ด, ์‹œ์Šคํ…œ์„ ํฌํ•จํ•œ ์—ฌ๋Ÿฌ ์—”์ง€๋‹ˆ์–ด๋ง ์˜์—ญ์—์„œ AI ๊ธฐ๋ฐ˜ ๋ฌธ์ œ ํ•ด๊ฒฐ ์—ญ๋Ÿ‰์„ ํ™•์žฅํ•˜๋Š” ๋ฐ ์ง‘์ค‘ํ•˜๊ณ  ์žˆ์Šต๋‹ˆ๋‹ค.

๋‹จ์ˆœํžˆ ๊ธฐ์กด ์ž‘์—…์„ ์ž๋™ํ™”ํ•˜๋Š” ๊ฒƒ์ด ์•„๋‹ˆ๋ผ, AI์™€ ์—”์ง€๋‹ˆ์–ด๊ฐ€ ํ•จ๊ป˜ ๋ฐ˜๋ณต์ ์ธ ํƒ์ƒ‰๊ณผ ์ตœ์ ํ™” ๊ณผ์ •์— ์ฐธ์—ฌํ•  ์ˆ˜ ์žˆ๋Š” ์ƒˆ๋กœ์šด ์ ‘๊ทผ ๋ฐฉ์‹์„ ๋งŒ๋“ค์–ด๊ฐ€๊ณ  ์žˆ์Šต๋‹ˆ๋‹ค. ์ด๋ฅผ ํ†ตํ•ด ๋ณด๋‹ค ๋‹ค์–‘ํ•œ ๊ฐ€๋Šฅ์„ฑ์„ ํƒ์ƒ‰ํ•˜๊ณ , ๋ฐ˜๋ณต์ ์ธ ๋ฌธ์ œ ํ•ด๊ฒฐ ๊ณผ์ •์„ ํšจ์œจ์ ์œผ๋กœ ์ˆ˜ํ–‰ํ•  ์ˆ˜ ์žˆ๋„๋ก ๋•๊ณ ์ž ํ•ฉ๋‹ˆ๋‹ค. ๊ถ๊ทน์ ์œผ๋กœ๋Š” ์—”์ง€๋‹ˆ์–ด๊ฐ€ ๋ฐ˜๋ณต์ ์ด๊ณ  ์†Œ๋ชจ์ ์ธ ์ž‘์—…์—์„œ ๋ฒ—์–ด๋‚˜ ๋” ์ค‘์š”ํ•œ ํŒ๋‹จ๊ณผ ์„ค๊ณ„, ์ฐฝ์˜์ ์ธ ๋ฌธ์ œ ํ•ด๊ฒฐ์— ์ง‘์ค‘ํ•  ์ˆ˜ ์žˆ๋Š” ๋ฐฉํ–ฅ์„ ์ง€ํ–ฅํ•˜๊ณ  ์žˆ์Šต๋‹ˆ๋‹ค.

About the Role

์ด ํฌ์ง€์…˜์€ AI๋ฅผ ํ™œ์šฉํ•ด ๋ณต์žกํ•˜๊ณ  ๋ฐ˜๋ณต์ ์ธ ํƒ์ƒ‰ยท๊ฐœ์„  ๊ณผ์ •์ด ํ•„์š”ํ•œ ์—”์ง€๋‹ˆ์–ด๋ง ๋ฌธ์ œ๋ฅผ ํ•ด๊ฒฐํ•  ์ˆ˜ ์žˆ๋Š” autonomous agent system์„ ๊ฐœ๋ฐœํ•˜๋Š” ์—ญํ• ์ž…๋‹ˆ๋‹ค.

Agentic AI๋ฅผ ๋‹จ์ˆœํ•œ LLM wrapper๋‚˜ workflow automation์œผ๋กœ ๋‹ค๋ฃจ๊ธฐ๋ณด๋‹ค, AI๊ฐ€ ๋ฌธ์ œ๋ฅผ ์ดํ•ดํ•˜๊ณ  ๋‹ค์–‘ํ•œ ํ•ด๊ฒฐ ๋ฐฉํ–ฅ์„ ์‹œ๋„ํ•˜๋ฉฐ ๊ทธ ๊ฒฐ๊ณผ๋ฅผ ๋ฐ”ํƒ•์œผ๋กœ ์Šค์Šค๋กœ ๊ฐœ์„ ํ•ด ๋‚˜๊ฐ€๋Š” ์‹ค์งˆ์ ์ธ ๋ฌธ์ œ ํ•ด๊ฒฐ ์‹œ์Šคํ…œ์œผ๋กœ ๊ตฌํ˜„ํ•˜๋Š” ๋ฐ ์ง‘์ค‘ํ•ฉ๋‹ˆ๋‹ค. ํŠนํžˆ ๊ณ ์ •๋œ ํ˜•ํƒœ์˜ agent system์ด ์•„๋‹ˆ๋ผ, ์‹คํ–‰ ๊ฒฐ๊ณผ์™€ ํ”ผ๋“œ๋ฐฑ์„ ๋ฐ”ํƒ•์œผ๋กœ ์ง€์†์ ์œผ๋กœ ์„ฑ๋Šฅ๊ณผ ๋™์ž‘ ๋ฐฉ์‹์ด ๋ฐœ์ „ํ•ด ๋‚˜๊ฐˆ ์ˆ˜ ์žˆ๋Š” self-improving system์„ ๋งŒ๋“œ๋Š” ๊ฒƒ์„ ์ค‘์š”ํ•œ ๋ฐฉํ–ฅ์œผ๋กœ ๋ณด๊ณ  ์žˆ์Šต๋‹ˆ๋‹ค.

์ด๋ฅผ ์œ„ํ•ด agent system์˜ ๊ตฌ์กฐ์™€ ๋™์ž‘ ๋ฐฉ์‹์„ ์„ค๊ณ„ํ•˜๊ณ , agent๊ฐ€ ์•ˆ์ •์ ์œผ๋กœ ๋ฌธ์ œ๋ฅผ ํƒ์ƒ‰ยท์‹คํ–‰ยทํ‰๊ฐ€ํ•  ์ˆ˜ ์žˆ๋Š” ๊ธฐ๋ฐ˜ ์‹œ์Šคํ…œ์„ ๊ฐœ๋ฐœํ•˜๊ฒŒ ๋ฉ๋‹ˆ๋‹ค. ๋˜ํ•œ ๋ฌธ์ œ์™€ ์‹œ์Šคํ…œ์˜ ํŠน์„ฑ์— ๋”ฐ๋ผ LLM post-training, agent system evaluation, ์‹ค์ œ ์—”์ง€๋‹ˆ์–ด๋ง workflow์™€์˜ integration ๋“ฑ์˜ ๊ณผ์ œ๋ฅผ ๋‹ค๋ฃจ๊ฒŒ ๋˜๋ฉฐ, ์ƒˆ๋กœ์šด ์•„์ด๋””์–ด๋ฅผ ์‹ค์ œ ๋™์ž‘ํ•˜๋Š” production-quality system์œผ๋กœ ๋ฐœ์ „์‹œํ‚ค๋Š” ๊ณผ์ •์— ์ฐธ์—ฌํ•˜๊ฒŒ ๋ฉ๋‹ˆ๋‹ค.

What You'll Work On

  • Agent planning, tool use, memory, reasoning, self-evolving mechanism ๋“ฑ agent behavior ์—ฐ๊ตฌ ๋ฐ ๊ตฌํ˜„

  • ์—ฌ๋Ÿฌ agent๊ฐ€ ํ˜‘์—…ํ•˜๋ฉฐ solution space๋ฅผ ํƒ์ƒ‰ํ•  ์ˆ˜ ์žˆ๋Š” multi-agent workflow ๋ฐ orchestration ๊ฐœ๋ฐœ

  • Agent๊ฐ€ ์ƒ์„ฑํ•œ ๊ฒฐ๊ณผ๋ฅผ ์ž๋™์œผ๋กœ ์‹คํ–‰ยท๊ฒ€์ฆยทํ‰๊ฐ€ํ•˜๊ณ , ์ด๋ฅผ ๋ฐ”ํƒ•์œผ๋กœ system improvement๋ฅผ ๋ฐ˜๋ณตํ•  ์ˆ˜ ์žˆ๋Š” evaluation system ๊ฐœ๋ฐœ

  • LLM-as-a-judge, task-specific metric, benchmark ๋“ฑ์„ ํ™œ์šฉํ•œ agent evaluation ๋ฐ optimization

  • SFT, RL ๋“ฑ ๋‹ค์–‘ํ•œ post-training ๊ธฐ๋ฒ• ์‹คํ—˜ ๋ฐ ์ ์šฉ

  • ๋น ๋ฅด๊ฒŒ ๋ณ€ํ™”ํ•˜๋Š” Agentic AI์™€ post-training ์—ฐ๊ตฌ ํ๋ฆ„์„ ์‹ค์ œ ์—”์ง€๋‹ˆ์–ด๋ง ๋ฌธ์ œ ํ•ด๊ฒฐ์— ์ ์šฉ ๊ฐ€๋Šฅํ•œ ํ˜•ํƒœ๋กœ ๊ตฌํ˜„

Minimum Qualifications

  • ์ปดํ“จํ„ฐ๊ณตํ•™ ๋˜๋Š” ์œ ๊ด€ ๋ถ„์•ผ ํ•™์‚ฌ ํ•™์œ„

  • PyTorch, JAX, TensorFlow ๋“ฑ ํ•˜๋‚˜ ์ด์ƒ์˜ ML framework๋ฅผ ํ™œ์šฉํ•œ ๋ชจ๋ธ ํ•™์Šต ๋˜๋Š” ์‹คํ—˜ ๊ฒฝํ—˜

  • LLM ๋˜๋Š” Agentic AI ๊ธฐ๋ฐ˜ ์‹œ์Šคํ…œ์„ ์ง์ ‘ ๊ตฌํ˜„ํ•˜๊ฑฐ๋‚˜ ์‹คํ—˜ํ•ด๋ณธ ๊ฒฝํ—˜

  • ๋…ผ๋ฌธ์ด๋‚˜ ์ตœ์‹  ์—ฐ๊ตฌ ๋‚ด์šฉ์„ ๋น ๋ฅด๊ฒŒ ์ดํ•ดํ•˜๊ณ  ์‹ค์ œ ๊ตฌํ˜„์œผ๋กœ ์—ฐ๊ฒฐํ•  ์ˆ˜ ์žˆ๋Š” ๋Šฅ๋ ฅ

Preferred Qualifications

  • ์ปดํ“จํ„ฐ๊ณตํ•™ ๋˜๋Š” ์œ ๊ด€ ๋ถ„์•ผ์˜ ์„์‚ฌ/๋ฐ•์‚ฌ ํ•™์œ„

  • Agent system ๊ฐœ๋ฐœ ๊ฒฝํ—˜: tool use, reasoning, memory, planning, multi-agent orchestration ๋“ฑ

  • LLM post-training ๊ฒฝํ—˜: SFT, RL, preference optimization ๋“ฑ

  • Agent evaluation system ๊ตฌ์ถ• ๊ฒฝํ—˜: benchmark design, LLM-as-a-judge, automated evaluation ๋“ฑ

  • LangGraph, AutoGen, Google ADK ๋“ฑ agent framework ์‚ฌ์šฉ ๋˜๋Š” ๊ฐœ๋ฐœ ๊ฒฝํ—˜

  • Distributed training, inference optimization, large-scale experimentation ๊ฒฝํ—˜

  • Agentic AI ๋ฐ post-training ์—ฐ๊ตฌ ๋‚ด์šฉ์„ ์‹ค์ œ ๋ฌธ์ œ ํ•ด๊ฒฐ์— ์ ์šฉํ•ด๋ณธ ๊ฒฝํ—˜

Contact

  • recruit@furiosa.ai

Frequently Asked Questions

Is the salary disclosed for the Machine Learning Engineer (Agentic AI) position at furiosa-ai?
The salary for this Machine Learning Engineer (Agentic AI) role at furiosa-ai is not publicly listed. Click "Apply Now" to learn more about the compensation package on their official careers page.
Where is the Machine Learning Engineer (Agentic AI) position at furiosa-ai located?
This Machine Learning Engineer (Agentic AI) role at furiosa-ai is based in Seoul HQ. The position is listed as on-site or hybrid. Check the full job description or apply directly to confirm the work arrangement.
Is the Machine Learning Engineer (Agentic AI) role at furiosa-ai full-time or part-time?
This is listed as a FullTime position. It is posted as a Machine Learning Engineer (Agentic AI) role in the Software department at furiosa-ai.
Which team or department does the Machine Learning Engineer (Agentic AI) at furiosa-ai belong to?
This Machine Learning Engineer (Agentic AI) position is part of the Software department at furiosa-ai. See the full job description for more information about the team structure and responsibilities.
How do I apply for the Machine Learning Engineer (Agentic AI) position at furiosa-ai?
Click the "Apply Now" button on this page. You will be redirected to furiosa-ai's official application portal hosted on ashby where you can submit your application directly.
When was the Machine Learning Engineer (Agentic AI) job at furiosa-ai posted?
This Machine Learning Engineer (Agentic AI) position at furiosa-ai was posted on May 19, 2026. Apply as soon as possible โ€” early applications are often reviewed first.
Machine Learning Engineer (Agentic AI)
furiosa-ai
Apply for this role โ†—

You'll be redirected to furiosa-ai's official application page on Ashby ATS.