Merge pull request #86 from anyscale/rag-dev-bootcamp

create a template for RAG Dev Bootcamp 2024

configs/rag-dev-bootcamp-mar2024/aws.yaml

@@ -0,0 +1,10 @@
+head_node_type:
+  name: head_node_type
+  instance_type: g5.4xlarge
+worker_node_types:
+- name: gpu_worker
+  instance_type: g5.4xlarge
+  min_workers: 1
+  max_workers: 1
+  use_spot: false
+auto_select_worker_config: true

configs/rag-dev-bootcamp-mar2024/gce.yaml

@@ -0,0 +1,11 @@
+# n1-standard-8-nvidia-t4-16gb-1  --> 8 CPUs, 1 GPU
+head_node_type:
+  name: head_node_type
+  instance_type: n1-standard-8-nvidia-t4-16gb-1
+worker_node_types:
+- name: gpu_worker
+  instance_type: n1-standard-8-nvidia-t4-16gb-1
+  min_workers: 1
+  max_workers: 1
+  use_spot: false
+auto_select_worker_config: true

templates/rag-dev-bootcamp-mar2024/README.md

@@ -0,0 +1,19 @@
+# RAG Developer Bootcamp Training Program
+
+Level up your coding skills at the RAG Developer Bootcamp!
+
+<img src="https://img.evbuc.com/https%3A%2F%2Fcdn.evbuc.com%2Fimages%2F693207549%2F1858223624353%2F1%2Foriginal.20240208-192428?w=940&auto=format%2Ccompress&q=75&sharp=10&rect=0%2C0%2C2160%2C1080&s=b6c48669e4026108c2a56859d5b38d1e">
+
+
+## Overview
+
+In recent months, Retrieval Augmented Generation or RAG became a central concept in the LLM apps development. Developers are using RAG to reduce hallucinations by grounding LLM output and adding context that wasn’t captured in the LLM training data.
+
+
+## Training Objectives
+
+The RAG Developer Bootcamp is a 3-part training program designed to help developers understand and implement RAG in their LLM apps. The training program will cover three main areas:
+- Part 1: RAG apps overview and quick-start with Canopy
+- Part 2: RAG Development: embeddings at scale
+- Part 3: RAG Development: evaluations
+