方舟服务端研发工程师（链路架构方向）-火山方舟
Core
Design and develop core components for the MaaS platform (Model as a Service) to support text dialogue, multimodal understanding, and multimodal generation APIs.
Role type
Senior IC backend engineer (MaaS platform & cloud-native architecture)
Builds
High-throughput, low-latency inference services for large language models and multimodal AI
Domain
Cloud-native infrastructure, distributed systems, large model inference
Deliverable
production ML models
Required skills
Golang/Java/C++/Python, distributed systems, microservices architecture, high-concurrency systems, caching, message queues, load balancing
Preferred skills
Agent concepts (Function Calling, MCP), K8s, Docker, Istio/Envoy, multimodal data processing, LLM inference engineering
Responsibilities
Design and develop core MaaS platform components; evolve Service Mesh and intelligent routing architectures; optimize system performance for long connections and streaming output; ensure system stability under sudden traffic spikes