moving from mikrotik 5009 router to computer turned router

mik POS crashed and bricked, 3 times the charm

This is a 6th gen intel optiplex slim pc conversion. ssd drive

Phase 1: Parallel setup (keep TP-Link as backup)

• Keep current setup working

• Test Dell router with just a few devices

• Make sure everything works perfectly

Phase 2: Switch WAN connection

• Move cable modem to Dell enp3s0

• Configure TP-Link as access point only (disable its DHCP)

• Dell becomes main router for 192.168.0.x network

Diode – semiconductor, low-cost engraving

CO₂ (DC Glass Tube) – midrange cutting, hobby to production

CO₂ (RF Metal Tube) – high-end engraving, precision cutting, industrial reliability

Fiber – metal marking/cutting

DPSS / Nd:YAG – precision metal marking, jewelry, semiconductors

They are a type of CO₂ laser, but instead of being DC-excited glass tubes (like most Chinese imports), RF tubes use radio-frequency excitation and metal or ceramic enclosures.
So while they share the same wavelength (10.6 μm infrared light), the construction and performance are completely different.

Laser  Buy American, don't be fooled by China Crap! 

https://forumla.us/2025/10/18/8613/

What Is Ollama?

**Ollama** allows you to:

- Download and run popular open-source LLMs (like Llama 3, Mistral, Phi-3, Gemma, etc.) directly on your laptop or local server.

- Run these models **locally**, meaning no data leaves your device.

- Interact with models via a simple CLI (command-line interface) or API.

- Customize and fine-tune models for specific tasks.

Think of it as a **local AI assistant toolkit**

 create a local API server that allows LM Studio to "learn" from your Chroma database, here's a complete solution:

## Complete RAG API Server for Learning

```python

from sentence_transformers import SentenceTransformer

import chromadb

from flask import Flask, request, jsonify

from flask_cors import CORS

import os

import json

semantic search is by using embedding models to convert text into numerical vectors, then storing these vectors in a vector database, and finally using a K-Nearest Neighbor (KNN) algorithm to find the most semantically similar vectors to a user's query. This process relies on Natural Language Processing (NLP) and machine learning to understand the meaning and context behind words, enabling more relevant and accurate search results than traditional keyword matching

The Build is complete, next the build out!

You're not just building an AI tool — you're building a **personal AI system** that:

- Runs locally (on your 3090 Ti)

- Uses your favorite model (Qwen3 30B)

- Stays up to date with **real-time data**

- Is scalable, maintainable, and won’t crash from "tokens"

- Uses **you.com** (your own AI-powered project) as the core

.