---
title: 'LLM Inference Engineer at LG AI Research'
canonical: 'https://feeny.ai/job/llm-inference-engineer-lg-ai-research-gangseo-gu-seoul-south-pvnck6897ems'
type: 'job'
last_seen: '2026-09-11'
---

# LLM Inference Engineer at LG AI Research

- **Company:** LG AI Research
- **Location:** Gangseo-Gu Seoul South, South Korea
- **Posted:** 2025-09-10
- **Last confirmed live:** 2026-09-11
- **Apply:** https://job-boards.greenhouse.io/lgairesearch/jobs/4607350005

## Job description

팀 소개

Platform&Infra Team은 AI모델의 개발부터 서비스 운영을 위한 배포에 이르기까지 AI모델의 수명 주기를 최적화하고, 효율적으로 관리하기 위한 MLOps 파이프라인을 구축합니다. 또한 AI서비스의 안정적인 운영 지원을 위한 보안성 강화, 인프라 관리 및 자원 최적화 업무를 수행합니다. 팀 내에서 LLM Inference Engineer는 대규모 언어 모델(LLM)의 추론 성능을 최적화하고, 다양한 프레임워크에서 EXAONE모델을 안정적으로 서빙할 수 있는 엔진을 개발합니다. 또한, 오픈소스 프로젝트의 공식 채널을 통해 요청되는 기능 개선, 버그 수정, 성능 최적화 등의 요건에 대응하며, 다양한 환경에서 LLM이 안정적으로 동작할 수 있도록 기여합니다.

수행 업무

- LLM 추론 성능 최적화: 대규모 언어 모델(LLM)의 추론 성능을 극대화하기 위한 개선 방안을 도출하고 설계에 참여합니다.
- 서빙 엔진 개발: 다양한 프레임워크 상에서 엑사원 모델을 안정적으로 서빙할 수 있는 엔진을 개발합니다.
- 오픈소스 기여: 오픈소스 LLM 서빙 프레임워크(vLLM, SGLang, Ollama 등)의 공식 채널을 통해 요청되는 기능 개선, 버그 수정, 성능 최적화 등의 요건에 대응하고 기여합니다.
- 다양한 환경 지원: chat_template을 활용하여 다양한 언어와 환경에서 LLM이 올바르게 동작할 수 있도록 적용하고 개선합니다.
- 모델 추론 가속화: NPU/GPU 기반의 Serving 전략을 설계하고, 모델 추론 가속화 알고리즘을 연구 및 구현합니다.

지원자격

- 모델 추론 가속화 및 Inference Back-end 구현 경험: 모델 추론 성능을 높이기 위한 알고리즘 연구 및 구현 경험
- NPU/GPU 기반 Serving 전략 및 개발 경험: NPU/GPU를 활용한 모델 서빙 전략 설계 및 구현 경험
- 오픈소스 LLM Serving Framework 기여 경험: vLLM, SGLang, Ollama 등 오픈소스 LLM 서빙 프레임워크에 기여한 경험(contribution)

*우대사항

- 머신러닝 모델 수명 주기에 대한 이해: 모델 개발, 학습, 서빙, 배포 등 각 단계에 대한 경험과 이해
- AI 및 머신러닝 기술에 대한 깊은 이해: Python, C++, TensorFlow, PyTorch, ONNX, CUDA, NPU 등 AI 및 머신러닝 관련 기술에 대한 숙련도
- Transformer 기반 모델 및 LLM에 대한 이해: Attention 메커니즘, 대규모 언어 모델(LLM)의 구조와 동작 원리에 대한 깊은 이해
- LLM 디코딩 전략 및 추론 기술에 대한 이해: Greedy Search, Beam Search, Top-k Sampling 등 디코딩 전략에 대한 이해
- chat_template 활용 경험: 다양한 환경에서 LLM이 올바르게 동작할 수 있도록 chat_template을 설계하고 적용한 경험

전형절차

- 서류심사 → 코딩테스트 → 1차  PT직무 인터뷰(온라인) → 2차  직무 인터뷰(온라인) → LG AI Fit Check(온라인) → 최종 합격
- 전형 절차는 변경될 수 있습니다. 서류 합격 시 전형 절차에 대해 별도로 안내 드립니다.

현재 LG AI연구원은 병역지정업체가 아니므로, 전문연구요원 채용 및 전직이 불가함을 알려드립니다. 지원시 문제가 있을 경우 careers@lgresearch.ai 로 문의 부탁드립니다.

## About LG AI Research

## Company Overview
- **One-liner**: LG AI Research is the corporate AI research lab of the LG Group, dedicated to solving complex real-world problems across LG’s diverse businesses through cutting-edge AI research in vision, language, data intelligence, and materials science.
- **Entity Type**: Private (Subsidiary of LG Group, a publicly traded conglomerate; LG Corp ticker: 003550 on KOSPI)
- **Headquarters**: Seoul, South Korea (multiple offices: Yeouido LG Twin Tower, Magok LG Science Park, Magok D&O)
- **Founded**: December 2020
- **Founders**: Established by LG Group; Chief Scientist: Honglak Lee (also Professor at University of Michigan)

## Core Business
- Primary industry/industries: AI Research & Development, Research Services
- Target customers: Primarily internal – LG Group affiliates (LG Electronics, LG Chem, LG Display, LG U+, etc.); also collaborates externally with academic institutions and publishes research.
- Mission or purpose statement: "Advancing AI for a better life" by solving the LG Group’s most challenging problems and conducting fundamental research for future breakthroughs.

## Products & Services
LG AI Research does not sell commercial products; it operates specialized research labs that deliver models, platforms, and intelligence capabilities to LG businesses and the broader AI community.

- **[EXAONE Lab]**: Large Language Model (LLM) research and development, including the EXAONE series of models.
- **[Vision Lab]**: Computer vision research for applications in manufacturing, quality inspection, autonomous systems, etc.
- **[Language Lab]**: Natural language processing, conversational AI, multilingual understanding.
- **[Data Intelligence Lab]**: Data-driven decision-making, predictive analytics, and business optimization.
- **[Materials Intelligence Lab]**: AI for new material discovery, chemistry, and energy research.
- **[Bio Intelligence Lab]**: AI applications in drug discovery, life sciences, and protein design.
- **[Advanced Agent Lab]**: Agentic AI, autonomous systems, and robotics evaluation & benchmarking.
- **[Physical Intelligence Lab]**: Embodied AI, robotics, and physical world interaction.
- **[Superintelligence Lab]**: Long-term fundamental research toward general intelligence.
- **[Platform & Infra Team]**: MLOps, LLM inference infrastructure, and internal AI platforms.
- **[AI Market Insights Team]**: Business intelligence and market analysis for AI strategy.

## Market Standing
- **Valuation/Market Cap**: Not disclosed (wholly owned by LG Group)
- **Key Metric**: Total headcount of 255 employees (as of mid-2025), with +23.7% YoY growth
- **Notable Investors/Partners**: LG Group affiliates (LG Electronics, LG Chem, LG Display, LG U+, etc.); academic partners including University of Michigan, KAIST, Seoul National University, Yonsei University.
- **Growth Signals**: Headcount growth of 23.7% YoY; expansion to global offices in Ann Arbor (USA), Poland, and Canada; active hiring across multiple research and engineering roles; hosting events like "LG AI Tech CONNECT" in New York; regular publications at top global conferences.

## Competitive Advantages
- **Real-world data & problems**: Unparalleled access to diverse, large-scale industrial data from LG’s electronics, chemical, telecom, battery, and life science businesses.
- **End-to-end impact**: Researchers see their work directly applied to actual business outcomes within LG affiliates.
- **Top-tier infrastructure**: State-of-the-art compute resources, extensive datasets, and latest equipment.
- **World-class talent**: Collaboration with leading academic institutions and a team of highly cited researchers.
- **Autonomous culture**: Culture 3.0 emphasizes researcher autonomy, flextime without core hours, self-approval PTO, and remote work options.

## Strategic Focus
- **Fundamental research**: Continued investment in foundational AI areas (LLMs, agentic AI, scientific foundation models, materials intelligence).
- **Global expansion**: Growing research presence in North America (Ann Arbor) and Europe.
- **Industry-specific AI**: Deepening AI capabilities for LG’s key sectors (chemicals, displays, batteries, new drugs).
- **Talent attraction**: Actively recruiting top AI researchers and engineers through internships, residencies/postdocs, and full-time positions.

## Why Work Here
- **Autonomous work environment**: Flextime with no core hours, self-approval for paid time off, remote work options, and commuter shuttle service.
- **Health & wellness**: Comprehensive medical expense coverage (self & family), annual physical exams, psychological counseling (6 sessions/year), health management allowance.
- **Personal growth**: Generous self-development budget, book allowance, latest work equipment (laptop, monitors), access to vast data and infrastructure.
- **Life support**: Welfare points, family event support, childcare expenses, admission gifts, resort/villa support, separate summer vacation days.
- **Culture**: "Culture 3.0" – a continuously evolving, researcher-centric culture with unlimited snack bars, subsidized meals, and a focus on work-life balance.
- **Global opportunities**: Offices in Seoul, Ann Arbor (USA), Poland, and Canada, with potential for international collaboration and relocation.

## Sources
1. [lgresearch.ai](https://www.lgresearch.ai/)
2. [linkedin.com](https://www.linkedin.com/company/lgairesearch/)
3. [greenhouse.io](https://job-boards.greenhouse.io/lgairesearch)
4. [careers.lgresearch.ai](https://careers.lgresearch.ai/c5a2b8bd-6129-4482-a68e-a7dc0d300676)

## Other roles at LG AI Research

- [Research Internship - Physical AI](https://feeny.ai/job/research-internship-physical-ai-lg-ai-research-gangseo-gu-seoul-south-sw26vgp1sfyc) — Gangseo-Gu Seoul South, South Korea
- [Data Engineer Internship - Physical AI](https://feeny.ai/job/data-engineer-internship-physical-ai-lg-ai-research-gangseo-gu-seoul-south-8rnk0axd97mf) — Gangseo-Gu Seoul South, South Korea
- [클라우드 보안 담당자](https://feeny.ai/job/lg-ai-research-gangseo-gu-seoul-south-1vhmg11mt26x) — Gangseo-Gu Seoul South, South Korea
- [AI Software Engineer - Self-driving Laboratory](https://feeny.ai/job/ai-software-engineer-self-driving-laboratory-lg-ai-research-gangseo-gu-seoul-35fzmsa40bt8) — Gangseo-Gu Seoul South, South Korea
- [Software Engineer - Backend](https://feeny.ai/job/software-engineer-backend-lg-ai-research-gangseo-gu-seoul-south-ak5sgeryghtz) — Gangseo-Gu Seoul South, South Korea
- [AI 사업개발 변호사(EXAONE Nexus)](https://feeny.ai/job/ai-exaone-nexus-lg-ai-research-gangseo-gu-seoul-south-4xrbfwg87qtx) — Gangseo-Gu Seoul South, South Korea
- [Research Scientist/Engineer - Agent Memory](https://feeny.ai/job/research-scientist-engineer-agent-memory-lg-ai-research-gangseo-gu-seoul-south-rj7g6pgvtfrp) — Gangseo-Gu Seoul South, South Korea
- [AI Safety & Policy Specialist](https://feeny.ai/job/ai-safety-policy-specialist-lg-ai-research-gangseo-gu-seoul-south-7heb5vxdzvyg) — Gangseo-Gu Seoul South, South Korea
- [Enterprise Customer Success Manager](https://feeny.ai/job/enterprise-customer-success-manager-lg-ai-research-gangseo-gu-seoul-south-1qkv53ccnp9r) — Gangseo-Gu Seoul South, South Korea
- [AI 사업개발 전문가(국방분야)](https://feeny.ai/job/ai-lg-ai-research-gangseo-gu-seoul-south-x7ybm7m9vvjr) — Gangseo-Gu Seoul South, South Korea