Principal ML Solutions Architect - Token Factory
AI summary of the role
Principal ML Solutions Architect for Nebius Token Factory, a serverless platform for running and customizing open-source LLMs in production.
What you’ll do
- Own the most complex, highest-stakes customer engagements from architecture through production across multiple modalities
- Optimize LLM inference at the framework and hardware level and codify best practices into reusable playbooks
- Lead supervised and reinforcement fine-tuning efforts to maximize model quality
- Design and implement production-ready LLM solutions using Token Factory's inference services
What you’ll bring
- 8+ years in ML/AI systems with at least 4 years focused on LLMs and generative AI
- Expert knowledge of LLM ecosystem: model architectures, fine-tuning approaches, inference internals
- Deep hands-on command of inference optimization: quantization, KV-cache management, batching, routing
- Hands-on experience with inference frameworks (vLLM, SGLang, TensorRT-LLM) including ability to modify internals
Technologies
vLLM · SGLang · TensorRT-LLM · LoRA · KV-cache · quantization · speculative decoding · RAG · Python · Kubernetes
About Nebius
Full-stack AI cloud infrastructure platform delivering GPU compute and software to hyperscalers and enterprises without requiring in-house AI/ML teams.
Public · 1000–2000 people
Employer postings · Data from · Sources