Principal ML Solutions Architect - Token Factory
Core
Senior technical authority for customers leveraging Nebius Token Factory's serverless inference and fine-tuning platforms, designing optimized workflows and setting technical direction for strategic accounts.
Role type
Principal ML Solutions Architect (LLM Inference & Fine-tuning)
Builds
Serverless LLM inference and fine-tuning workflows for enterprise customers
Domain
Cloud Infrastructure / Generative AI / LLM Serving
Deliverable
production ML models
Required skills
LLM inference optimization, fine-tuning (SFT/LoRA/RL), prompt engineering, RAG architectures, Python programming, inference frameworks (vLLM, SGLang, TensorRT-LLM), model evaluation, cloud platforms (AWS/GCP/Azure)
Preferred skills
OSS contributions to inference projects, published research, multimodal AI expertise, DevOps tooling (Kubernetes, Docker)
Technologies
vLLM, SGLang, TensorRT-LLM, Transformers, OpenAI/Anthropic SDKs, Kubernetes, Docker, Git, AWS SageMaker/Bedrock, GCP Vertex AI, Azure ML
Responsibilities
Own complex customer engagements from architecture to production, optimize LLM inference at framework/hardware level, lead supervised and reinforcement fine-tuning efforts, design production-ready LLM solutions, partner with product/engineering to influence roadmap, mentor Solutions Architects, represent Token Factory externally
Seniority
Principal, hands-on IC with strategic leadership
