Machine Learning Engineer

리스크엑스는 은행, 증권사, 자본시장 참여기관을 위한 AI 네이티브 플랫폼으로 구조화상품 시장의 혁신을 이끄는 핀테크 SaaS 기업입니다.

리스크엑스는 구조화상품 라이프사이클 전반의 핵심 업무를 자동화함으로써, 금융기관이 수작업 중심의 업무를 줄이고, 견적 산출 시간을 단축하며, 리스크 관리 및 감사 대응 체계를 고도화할 수 있도록 지원합니다.

플랫폼은 API 기반으로 실시간 파생상품 가격 산출 및 리스크 지표를 제공하여, 유통 및 평가 업무의 자동화와 확장성을 가능하게 합니다.

리스크엑스는 의사결정 자동화 영역으로 역량을 지속적으로 확장하며, 고도화된 시장 시나리오 시뮬레이션, 스트레스 테스트, 기관 요구사항에 부합하는 시나리오 기반 추천 기능을 지원하고 있습니다.

리스크엑스는 AI 전환(AX)을 통해 금융기관이 업무 효율성, 투명성, 투자 성과를 높이고, 기존 프로세스를 현대화할 수 있도록 돕고 있습니다.

RiskX is a fintech SaaS company revolutionizing the structured products space with an AI-native platform designed for banks, securities firms, and capital markets. By automating critical aspects of the structured product lifecycle, RiskX helps institutions reduce manual workflows, accelerate time-to-quote, and enhance risk management and audit readiness. The platform delivers real-time derivatives pricing and risk metrics through APIs, enabling scalable and automated solutions for distribution and valuation. Committed to innovation, RiskX continues to expand its capabilities toward decision automation, supporting advanced market-scenario simulations, stress testing, and scenario-based recommendations aligned with institutional requirements. RiskX empowers financial institutions to modernize their processes, improving efficiency, transparency, and investment outcomes through AI transformation (AX).

포지션 소개

리스크엑스는 금융상품 추천시스템의 핵심 추천 파이프라인을 구축하고 확장할 ML 엔지니어를 찾고 있습니다.

이 역할에서는 상품 속성, 시장 데이터, 시뮬레이션 결과, 투자자 프로필, 포트폴리오 맥락을 통합하여 개인화된 추천 모델을 설계하고 고도화하게 됩니다.

또한 데이터 생성과 라벨 설계부터 모델 개발, 오프라인 평가, 프로덕션 서빙까지 시스템의 전 생애주기에 걸쳐 업무를 수행하게 됩니다.

복잡하고 불확실성이 높은 금융 문제를 견고한 머신러닝 시스템으로 구조화하고, 연구 결과를 실제 제품 성과로 연결하는 데 흥미를 느끼는 분께 적합한 포지션입니다.

About the Role

We are looking for an ML Engineer to build and scale the core recommendation pipeline for our financial product recommendation system.

In this role, you will design and improve personalized recommendation models by integrating product attributes, market data, simulation outputs, investor profiles, and portfolio context. You will work across the full lifecycle of the system, from data generation and label design to model development, offline evaluation, and production serving.

This position is well suited for someone who enjoys turning complex, ambiguous financial problems into robust machine learning systems and translating research into measurable product impact.

아래와 같은 직무에 관심있는 많은 분들의 지원을 희망합니다.

What You Will Do

• Research and develop recommendation models using Learning-to-Rank, representation learning, and reinforcement learning approaches
• Build personalized recommendation models that reflect investor characteristics, market conditions, product attributes, and portfolio context
• Improve recommendation performance across retrieval, ranking, and reranking stages
• Design conditional recommendation architectures based on investor risk tolerance, investment objectives, existing positions, and suitability constraints
• Develop feature engineering pipelines using product attributes, payoff-related characteristics, market data, and time-series statistics
• Build simulation-driven data generation pipelines using Monte Carlo methods and scenario-based modeling
• Design training labels based on realized outcomes, investor decisions, and holding or redemption behavior
• Evaluate recommendation quality through offline testing, backtesting, and failure case analysis
• Deploy models into production and improve serving performance, monitoring, and reliability

Qualifications

• Master’s degree or higher in Computer Science, Statistics, Industrial Engineering, Financial Engineering, or a related field
• Hands-on experience with model training and inference using PyTorch
• Solid understanding of recommendation systems, ranking models, or personalization
• Strong skills in data preprocessing, experiment design, result analysis, and model debugging
• Experience writing reproducible code and managing ML experiments
• Ability to read research papers or technical documents and translate key ideas into working implementations
• Ability to formulate ambiguous real-world problems into structured ML problems

Preferred Qualifications

• Domain knowledge in derivatives, quantitative finance, option pricing, or investment product design
• Experience with Monte Carlo simulation, stochastic processes, or scenario generation
• Practical experience in recommendation systems, including candidate generation, ranking, reranking, two-tower retrieval, negative sampling, and bias mitigation
• Experience with financial time-series modeling or generative models such as GANs, diffusion models, or flow matching
• Experience serving recommendation systems in production, including model deployment, feature stores, real-time inference, monitoring, and performance tracking
• Experience with reinforcement learning, contextual bandits, or sequential decision-making
• Experience with representation learning or LLM-related tools such as HuggingFace or SentenceTransformer
• Familiarity with ML pipeline and experimentation tools such as Spark, Airflow, Ray, MLflow, or Weights & Biases
• Publications, competition awards, or open-source contributions in relevant fields

Who We’re Looking For

• Someone who can define and solve open-ended problems through structured experimentation
• Someone who can work within financial domain constraints while reframing problems from an ML perspective
• Someone who can design not only models, but also labels, features, and evaluation frameworks
• Someone who can quickly absorb research ideas and compare experimental outcomes rigorously
• Someone who is excited about taking research prototypes into production systems

What You Will Own

• Ramp up quickly on the current recommendation pipeline, data structure, and key performance bottlenecks
• Propose and validate improved label, feature, and model designs for financial product recommendation
• Take ownership of the core recommendation engine over time, balancing performance, suitability, robustness, and scalability

What You’ll Gain

• The opportunity to work on a high-difficulty recommendation problem beyond conventional commerce or content recommendation
• End-to-end experience across quantitative modeling, simulation, recommendation system design, deployment, and production optimization
• Hands-on exposure to applied ML systems that connect research output to real product impact
• Experience integrating tabular data, time-series data, simulation outputs, and investor behavior data into a unified ML system