Machine Learning Engineer (Contents Understanding) at Match Group

Seoul, South Korea

Apply Now

Not SpecifiedCompensation

Mid-level (3 to 4 years), Senior (5 to 8 years)Experience Level

Full TimeJob Type

UnknownVisa

Technology, Social NetworkingIndustries

Requirements

Broad understanding of AI/ML domains and deep knowledge in at least one specific domain
Strong interest in productizing AI technologies
Proven capability to solve problems using AI, including project management skills
Strong Python development skills, including experience with open-source frameworks like TensorFlow, PyTorch, CatBoost, JAX
Strong ownership to lead and complete projects from A to Z
Excellent communication skills for collaborating with diverse stakeholders across teams
Engineering expertise to understand ML system software architecture and plan features
Ability to identify statistical characteristics and patterns in data and apply them to AI problem-solving
Experience planning A/B tests, defining target KPI metrics, and conducting SQL-based data analysis
Fluent communication in Korean (degree and nationality irrelevant)

Responsibilities

Research and improve state-of-the-art models while considering real-world constraints for engineering decisions
Identify product issues, reframe them as AI problems, and drive solutions end-to-end
Lead the full problem-solving lifecycle: problem definition, stakeholder alignment, goal setting, SOTA model derivation, scheduling, performance analysis, and future strategy
Collaborate closely with backend/frontend/DevOps/ML engineers, data analysts, PMs, and other teams
Define priorities based on user needs and business impact to drive long-term product growth via AI
Extract actionable insights from unstructured data (video, images, audio, natural language) to support content moderation, Trust & Safety, and global standards
Design lightweight models for mobile/web with low latency, low power, and high accuracy
Manage label quality in noisy/imbalanced data using active learning, core-set selection, semi-/self-supervised methods
Optimize multi-task/multi-label classification within parameter constraints and integrate multi-modal data (text, image, video)
Apply domain adaptation and meta-learning for cross-domain challenges and service scalability
Implement fairness, privacy-preserving learning to meet international AI standards
Develop streaming models for real-time anomaly detection (e.g., spam, fake accounts) using user logs and content analysis
Leverage LLMs to innovate ML production processes
Utilize on-premise GPU clusters (A100/H100) and cloud data pipelines for model development and deployment

Skills

Machine Learning

Deep Learning

Computer Vision

Natural Language Processing

Audio Processing

Multi-Modal Learning

Active Learning

Self-Supervised Learning

Semi-Supervised Learning

Multi-Task Learning

Multi-Label Classification

Domain Adaptation

Meta-Learning

Model Optimization

Low-Latency Inference

Data Pipelines

Cloud Services

Match Group

Provides online dating and social discovery

About Match Group

Match Group leverages the Swipe feature® and social discovery, facilitating deeper connections through its portfolio of online dating brands, with a global presence and availability in over 40 languages.

Dallas, TexasHeadquarters

1986Year Founded

$400MTotal Funding

IPOCompany Stage

Consumer Software, Social ImpactIndustries

1,001-5,000Employees