MaaS 架构师
Tasks
- Apply prefix caching and paged attention
- Build LLM inference gateway
- Configure priority queues and multi tenant isolation
- Create unified inference load abstraction
- Deploy quantized models
- Design MaaS platform architecture
- Design token level metering
- Develop GPU elastic scaling strategy
- Enable continuous batching and speculative decoding
- Ensure model service SLAs latency throughput
- Implement KV cache aware scheduling
- Implement model routing and traffic scheduling
- Package inference services with observability
- Select and optimize inference engine frameworks
Perks/Benefits
- N/A
Skills/Tech-stack
Attention | Batching | C++ | CUDA | Continuous batching | GPU scheduling | Go | KV cache | Kubernetes | LLM serving | Multi-tenancy | NCCL | Paged Attention | Prefix caching | Python | Quantization | Ray Serve | SGLang | Speculative decoding | TensorRT-LLM | Triton | VLLM
Education
Bachelor of Engineering | Bachelor of Science | Master of Science | PhD
Roles
Regions
Countries
States
Related jobs
-
Ai算法工程师 CNY 144K-240KApache Hadoop | Apache Hive | Apache Spark | Deep learning | Feature EngineeringEntry-level Full Time深圳1d ago
-
Agentic Engineering | Apache Airflow | Apache NiFi | Automated testing | BigQueryMid-level Full TimeGuangzhou, Guangdong, China2d ago
-
Apache Airflow | Apache Kafka | Apache NiFi | Automated testing | BigQueryMid-level Full TimeGuangzhou, GD, CN, 5106202d ago
-
Mid-level Full TimeShanghai Jing'An Office, China2d ago
-
Quantitative Researcher (AI Direction) CNY 144K-240KAlpha generation | Deep learning | High Frequency | High-Frequency Trading | JAXCareer development opportunities | Casual working environment | Digital asset market exposure | Meritocratic cultureSenior-level Full TimeShanghai3d ago
-
AI Data Scientist Intern CNY 38K-50KAI Algorithms | Data Annotation | Data Integrity | Data Modeling | Data QualityOnsite work at company facilitiesEntry-level Internship Part TimeShanghai - Daning A2 PIC, China3d ago
-
Mid-level Full TimeCN-OCG International Center, Cheng Du, China3d ago
-
Mid-level Full TimeCN-OCG International Center, Cheng Du, China3d ago
-
Senior-level Full Time深圳3d ago
-
Entry-level Internship深圳3d ago
-
Entry-level Internship上海3d ago
-
Senior Manager-AI Algorithm CNY 144K-240KAPI Development | Agent Orchestration | Agentic AI | CI/CD | Cloud PlatformsSenior-level Full TimeChina4d ago
-
Senior-level Full TimeChina4d ago
-
Engineering Team Lead – AI Applications (Life Sciences) CNY 360K-510KAgent systems | Agentic Workflows | Agile | Autogen | Backend architectureSenior-level Full TimeSuzhou5d ago
-
Jr. Digital Product Architect CNY 38K-50KAWS | AWS Glue | Agile | Alteryx | AzureMentoring support | Remote collaboration | Training opportunitiesEntry-level Full TimePudong New Area, SH, CN, 2012105d ago
-
ASIC | Agent systems | Agentic AI | BF16 | Bandwidth OptimizationSenior-level Full TimeShanghai, China5d ago
-
Mid-level Full Time深圳5d ago
-
Entry-level Full Time北京5d ago
-
Senior-level Full Time上海、杭州6d ago
-
Mid-level Full Time深圳、上海6d ago
-
Ai应用开发工程师（Pc方向） CNY 144K-240KCost Optimization | Data cleaning | Embedding | Engineering Automation | EvaluationEntry-level Full Time上海6d ago
-
【2026届春招】AI Infra工程师-算法研发方向 CNY 144K-240KC++ | CUDA | Data parallelism | Distributed Training | HuggingfaceEntry-level Full Time北京、上海 R8d ago
-
Mid-level Full Time北京8d ago
-
【校招实习生】AI Infra 工程师-算法研究方向 CNY 25K-37KC++ | CI/CD | CUDA | NCCL | OpenMPIFull-time internship | Open source project opportunitiesEntry-level Internship北京、上海 R8d ago
-
【27届校招】Omni模型算法工程师/研究员 CNY 500K-500KData Engineering | Distributed Training | Inference acceleration | Instruction Tuning | Knowledge DistillationEntry-level Full Time深圳、上海、北京9d ago