Staff Software Engineer, AI Model Lifecycle

Crusoe · Data AI · San Francisco, CA - US · Cloud Engineering

Staff Software Engineer focused on building a managed platform for the AI model lifecycle, including fine-tuning, training pipelines, and dataset/model management for LLMs and multimodal models.

What you'd actually do

  1. Manage fine-tuning systems for large foundation models (SFT, PEFT, LoRA, adapters), including multi-node orchestration, checkpointing, failure recovery, and cost-efficient scaling.
  2. Implement and maintain end-to-end training pipelines for Large Language Models.
  3. RFT and Reinforcement learning to the fine tuning and training sections
  4. Distillation and reinforcement learning pipelines (e.g., preference optimization, policy optimization, reward modeling).
  5. Dataset, model, and experiment management: versioning, lineage, evaluation, and reproducible fine-tuning at scale.

Skills

Required

  • Generative AI (Large Language Models, Multimodal)
  • training LLMs
  • fine-tuning LLMs
  • aligning LLMs
  • Reinforcement Learning
  • Reinforcement Fine-Tuning (RFT)
  • dataset management
  • model management
  • experiment management
  • multi-node orchestration
  • checkpointing
  • failure recovery
  • cost-efficient scaling
  • preference optimization
  • policy optimization
  • reward modeling
  • SFT
  • PEFT
  • LoRA
  • adapters

Nice to have

  • Golang
  • Python
  • PyTorch
  • vLLM
  • performance optimizations on GPU systems
  • inference frameworks

What the JD emphasized

  • 8+ years of industry experience leading and driving impactful projects in the AI Space
  • Experience in Generative AI (Large Language Models, Multimodal)
  • Hands-on experience training, fine-tuning, and aligning LLMs using Reinforcement Learning and Reinforcement Fine-Tuning (RFT) techniques.

Other signals

  • building managed platform for AI application development lifecycle
  • fine-tuning systems for large foundation models
  • end-to-end training pipelines for LLMs
  • dataset, model, and experiment management