LLM System Backend Engineer Graduate (AML Ark) - 2027 Start
Summary
Develops reinforcement learning infrastructure, RL training APIs, and optimizes LLM inference performance/cost for a multi-tenant, serverless platform.
- Build reinforcement learning infrastructure and RL training APIs
- Design elastic training solutions for multi tenant workloads
- Develop Model as a Service inference platform
- Develop training platform for serverless post training
- Optimize large language model inference performance cost and reliability
- Reduce inference costs with disaggregated inference distributed KV cache and elastic scheduling