njemanzedavid

The "Unified Neon + R2" Architecture

This model consolidates all structured and vector data into Neon, while dedicating Cloudflare R2 to what it does best: storing large, unstructured files.

Component	Service Used	Role in the Stack	Cost
1. Core n8n App	Oracle Cloud Free Tier	Your backend eg. the always-on n8n engine.	$0
2. Networking	Cloudflare Tunnel, DNS, WAF	The secure front door to your app.	$0
3. Unified Database	Neon Serverless Postgres	Stores application data (logs, credentials) AND your vector store (embeddings + metadata).	$0
4. Raw File Storage	Cloudflare R2	Stores all large source files (videos, audio, PDFs, etc.).	$0

	# Project Policy

	This policy provides a single, authoritative, and machine-readable source of truth for AI coding agents and humans, ensuring that all work is governed by clear, unambiguous rules and workflows. It aims to eliminate ambiguity, reduce supervision needs, and facilitate automation while maintaining accountability and compliance with best practices.

	# 1. Introduction

	> Rationale: Sets the context, actors, and compliance requirements for the policy, ensuring all participants understand their roles and responsibilities.

	## 1.1 Actors

	# train_grpo.py
	import re
	import torch
	from datasets import load_dataset, Dataset
	from transformers import AutoTokenizer, AutoModelForCausalLM
	from peft import LoraConfig
	from trl import GRPOConfig, GRPOTrainer

	# Load and prep dataset

	import cv2
	import numpy as np
	import mediapipe as mp

	# Initialize mediapipe Face Mesh and Hands solutions
	mp_face_mesh = mp.solutions.face_mesh
	mp_hands = mp.solutions.hands
	mp_drawing = mp.solutions.drawing_utils
	mp_drawing_styles = mp.solutions.drawing_styles