https://tech.instacart.com/introducing-griffin-2-0-instacarts-next-gen-ml-platform-b7331e73b8d7

Notes:

• V2 platform to offer new and emerging capabilities such as distributed training and fine-tuning of Large Language Models (LLMs).
  • BentoLM, Instacart’s in-house ML notebook cloud-based development environment
• three major subsystems
  • The model training platform (MLTP) leverages Ray to offer a horizontally scalable computing environment that supports the distributed training of ML models.
  • model service platform (MLSP) provides streamlined and automated model artifact storage, model deployments, and provisioning of inference services
  • The feature store supports feature computation, ingestion, discoverability, and shareability