Inference Framework Engineer for High-Scale AI

ByteDance; Singapore; On-site; SGD 120,000 - 180,000; Full time

Job summary

ByteDance is seeking talented engineers for the Data AML team. You will join a machine learning mid-platform that powers recommendations and ads for Douyin, Jinri Toutiao, and Xigua Video, delivering scalable inference services for large models.

The role offers opportunities to tackle challenging problems and grow with a global tech leader. Successful candidates should be strong in C/C++, Linux, and high-concurrency systems, and be eager to contribute to architecture design, performance

Qualifications

• Bachelor's or master's degree in computing or a related discipline.
• Proficient in Linux, C/C++, data structures and algorithms.
• Strong knowledge of multi-threading, locks, and thread pools.
• Experience with high-concurrency distributed services and latency optimization.
• Willing to understand model inference service architecture and provide problem analysis.
• Excellent cross-team collaboration, communication, and documentation skills.

Responsibilities

• Design and implement high-performance, scalable model inference services.
• Develop and optimize core modules: scheduling, monitoring, canary releases.
• Track latest inference tech and propose architecture improvements.

Skills

C/C++ programming

Linux commands

Multi-threading

Performance tuning

High-concurrency

Problem analysis

Cross-team collaboration

Education

Bachelor's or Master's in Computer Science/Computing

Tools

Redis

RocksDB

BRPC

gRPC

GPU service management

Job description

ByteDance is seeking talented engineers for the Data AML team. You will join a machine learning mid-platform that powers recommendations and ads for Douyin, Jinri Toutiao, and Xigua Video, delivering scalable inference services for large models.

The role offers opportunities to tackle challenging problems and grow with a global tech leader. Successful candidates should be strong in C/C++, Linux, and high-concurrency systems, and be eager to contribute to architecture design, performance