Introduction to Machine Learning Software Engineer

Hyperconnect Machine Learning Software Engineers (MLSE) apply machine learning technology to services that connect people through software engineering, innovating user experience. Their goal is to apply numerous models across various domains, including video, audio, text, and recommendations, created by the in-house AI organization, to production, provide them stably via mobile and cloud servers, solve engineering problems encountered, and ensure that the technology we create contributes to the growth of actual services.

Under this goal, Hyperconnect's ML Software Engineers have been developing machine learning technologies that contribute to various Hyperconnect products such as Azar and new in-house services for several years. They are also developing technologies to easily utilize these accumulated technologies in various global business services of Match Group, an S&P 500 company.

ML Software Engineers aim to apply all AI technologies we possess to products to create business impact and develop sustainable systems to accelerate the application of AI technology. To achieve this goal, they (1) develop scalable backend servers based on ML models, (2) develop/operate real-time data pipelines for ML model inference, (3) develop systems to support various operations involved in ML model production and operation, and (4) collaborate with other teams to develop a common platform that can be managed. In this process, they work closely with other roles and proactively carry out tasks, participating in all necessary processes (problem definition, hypothesis setting, experiment design, analysis, and feedback) to achieve KPI targets. For more details on how they work, please refer to the following:

• AI Platform Dev Team Interview
• ML Application Team Interview
• Head of AI - Shurain Interview
• AI in Social Discovery (Blending Research and Production)

The interesting problems ML Software Engineers are solving are uploaded on the Tech blog.

• Calculating Real-time ML Feature Data using ksqlDB
• 10 Python Performance Optimization Tips for High-Performance ML Backends
• Reducing Machine Learning Model Serving Costs by 1/4
• Operating an Event-Driven Live Streaming Recommendation System
• Data Storage Technologies for Machine Learning Applications

Work Environment

Top-tier AI Organization in Korea

You will work with Machine Learning Engineers and Machine Learning Research Scientists who regularly publish papers at top-tier AI/ML conferences. You can check out papers published by Hyperconnect here.

Rich MLOps Know-how

You can work in an organization with rich MLOps know-how, using over 50 models in production.

• Deview 2023 - Implementing a Feature Store for Real-time Recommendation Systems
• Deview 2021 - How to Deploy More Models Faster?
• Building a High-Performance Deep Learning Cluster Part 1

Responsibilities

• Hyperconnect is making various efforts to apply machine learning technology to its products. Hyperconnect's ML Software Engineer will primarily perform the following tasks:

Recommendation System Development

• Develop a personalized recommendation system for an enjoyable experience in Azar's core 1:1 video call feature. Also focus on developing recommendation and search systems in various other areas. These microservices, operated by the team, are designed with significant consideration for performance to operate in real-time on a global scale and handle some of the highest traffic within the company.
• Additionally, develop a real-time data pipeline for real-time model inference, processing real-time events using Apache Flink and KSQL, and providing data sources for models to utilize. Design systems (e.g., streaming applications, Feature Store) for fast and stable collection, processing, and serving of features, and also perform tasks to discover new features that can improve model performance.

Moderation System Development

• Develop moderation systems and platforms as part of Trust & Safety for a safe environment for customers. Build AI-based content and behavior monitoring systems to effectively detect inappropriate content and spammers.
• Furthermore, to continuously improve the performance of moderation models, develop platforms for producing and managing labeling data. Build Golden Data Sets to provide high-quality data for model training, and also design features to incorporate Human Factors instead of AI when needed for the service. This supports AI models in maintaining a reliable level of quality while solving various layers of moderation problems.

AI Feature Development Acceleration Strategy Support

• Build an AI Flywheel system to provide various functions for hypothesis testing, model training (continuous learning), evaluation, management, and deployment.

Requirements

• Fearless in speaking and reading English, capable of handling daily tasks in English.
• 3+ years of Software Engineering experience or equivalent skill.
• Solid foundational knowledge of CS fundamentals (operating systems, computer system architecture, data structures, and algorithms).
• Proficient in one or more of Java, Kotlin, Golang, Python, Javascript (Typescript), and able to quickly learn and use other programming languages.
• Understanding of various NoSQL databases and experience designing services using RDBMS.
• Proficient in using SQL for data exploration and capable of understanding the meaning of investigated data to use as evidence for problem-solving.
• Enjoys traversing various functional stacks and can quickly adapt to unfamiliar environments.
• Possesses strong communication skills to collaborate with stakeholders from various job functions.

Preferred Qualifications

• Enjoys researching new technologies or tackling challenging tasks.
• Experience using frameworks like FastAPI, SpringBoot, NestJS.
• Experience with open source contributions and managing forked repositories.
• Understanding and experience applying asynchronous network frameworks and technologies.
• Interest in MSA-based system architecture design.
• Understanding and experience applying the advantages and design benefits of Event Sourcing / CQRS.
• Experience from machine learning model training to service deployment.
• Experience integrating AI technology into actual services and significantly improving key metrics.
• Experience leading an engineering team.
• Experience developing backend servers that handle large-scale traffic.
• Experience developing distributed/parallel systems for large-scale/real-time processing such as Spark, Flink.

Employment Type/Recruitment Process

• Employment Type: Full-time
• Recruitment Process: Document Screening > Coding Test/Assignment > 1st Interview > Recruiter Call > 2nd Interview > Final Offer (*Process may change if necessary.)
• For document screening, only successful candidates will be notified individually.
• Application Documents: Free-form detailed English resume based on career (PDF)
• This position is available for Specialized Research Personnel (Hyunyeok/Bochungyeok) transfer/incorporation. For military service exemption personnel, service management will be conducted according to military service exemption laws.