TIONE-AI智能网关高级后台研发工程师-上海/杭州/深圳/北京
Core
Design and develop the core architecture for Tencent Cloud's TIONE inference platform and MaaS AI scheduling system, optimizing high-throughput, low-latency model inference services.
Role type
Senior IC backend engineer (AI infrastructure & MaaS)
Builds
High-performance model inference services, AI scheduling systems, and API gateways for MaaS products.
Domain
Cloud computing, AI infrastructure, Large Language Models (LLM)
Deliverable
production ML models
Required skills
Golang/Java/Python, high-concurrency distributed systems, Kubernetes, Docker, microservices, RPC/HTTP, message queues, GPU optimization, heterogeneous compute management, model quantization/distillation, vector databases
Preferred skills
Cloud provider or middleware R&D experience, MaaS/LLM inference service development, large-scale model service operations, AI security compliance, enterprise private deployment
Technologies
K8s, Docker, Golang, Java, Python, Kubernetes, Docker, microservices, RPC, HTTP, message queues, vector databases
Responsibilities
Develop core architecture for the TIONE inference platform; Design and develop the MaaS AI scheduling system and API gateway; Collaborate with algorithm and product teams to adapt models and optimize performance; Track industry trends in LLM engineering and cloud-native technologies to optimize architecture.
