Langchain PDF Ollama RAG (Retrieval Augmented Generation)

Give PDF and talk about it using lanchain to perform RAG using Ollama embedings and FAISS vectorstore yeah maybe many tools already provide this kind of feature, But this is different, this is about knowing behind the scene. what actually they do to the pdf files? what actually we do to the text in it? and how the LLM is know what context they need?. first thing first what is RAG? RAG is Retrieval Augmented Generation ...

2 December 2024 | prima adi

Ollama Chat From Browser Using Nextjs

Ollama Chat From Browser Using Nextjs This writing will be so small. it should be a step by step on creating nextjs app that will communicate with local ollama. the model that i use is mistral its the fastest now running in my local. but instead starting from scrath, lets asking help to bolt.new to create interface that will communicate with our ollama server in local. there is a reason why i learn this. i want to do …. lets wait for another post 😄 ...

21 November 2024 | prima adi

Running Qwen in M2 Mac Air Machine using Ollama

Running qwen-2.5-coder:7B on macbook Air M2 using ollama This is part of stay hungry stay folish mindset, and my interest in AI. Search possible solution to get cheapest code assistant as possible. i was subscribe to copilot but now after found continue.dev + anthropic API, thats the current choice. Its small decrease in cost. copilot is $10 per month but now i am not that code heavy so subs to token based payment (anthropic claude 3.5 sonnet) is more cost effective. and pay base on what i use. ...

20 November 2024 | prima adi

Getting Pull Rate Limit and Workaround

When docker hub want you to authenticate there is a stable system using public images from docker hub. and suddently pull rate limit error is showing up. This is How we implement the Workaround Our system is running in AWS EKS and using all internal image registry (ECR) for application images. Most of application is using docker images and run’s on kubernetes. but some POC and small services is using public images from docker hub. it needs you to authenticate or login to be able to increase the limit. ...

13 December 2022 | prima adi

ChatGPT Open Ai, Is It Scary?

ChatGPT is it scary? it could be the end of yaml (kubernetes) engineers. there is phenomenon. in this time i write this post, i just try the viral AI in a programming world. openAI just released a product, or reasearch, or a robot called chatGPT they said We’ve trained a model called ChatGPT which interacts in a conversational way. The dialogue format makes it possible for ChatGPT to answer followup questions, admit its mistakes, challenge incorrect premises, and reject inappropriate requests. ChatGPT is a sibling model to InstructGPT, which is trained to follow an instruction in a prompt and provide a detailed response. ...

9 December 2022 | prima adi

Argocd Starting Up

Argo CD Starting up Recently argocd is graduated from CNCF incubating project. we already using argocd in production almost 9 months. and its working awesome. Installation getting Started to install argocd and setting this up you could easy follow this (argocd start)[https://argo-cd.readthedocs.io/en/stable/getting_started/] Lets Go I will not explain how to install or how to configure argocd but i will share my experience of using it in production in our usecase. ...

30 November 2022 | prima adi

Layoff Phenomenon 2022

Layoff 2022 In the end of the 2022 there is phenomenon called “massive layoff”, a lot of my friends is impacted during this time. i just wonder what is happening. First, i am not an expert in investing, bussiness. but i kind of know (correct me if i am wrong) how to count a profit. its as simple as income - expenses so why the company as big as GoTo or RuangGuru is layoff their employee. i have hypotesis by my stupid logic. ...

25 November 2022 | prima adi

Hugo Is Calm Blog Engine

Hugo x Github Pages This is my 4th time to redeploy my blog as a note for myself. deciding using hugo x ghpages because it will be just a note could be a tech note or just random thought. why hugo is calm?, because its quite simple, its static, it is fun adn quite fast (to generate). Hugo what is hugo? Hugo is blog engine that use markdown as their main source of content. generate that markdown file as static site and deploy it. you could see more about hugo here https://gohugo.io/ ...

24 November 2022 | prima adi

About Teraskula

About i am prima wong biasa, kawulane gusti Allah, anake bapak ibuk, bapake anak anak, bojone bojoku. Teraskula.com is a new site that hopefully become my tool to release what is in my head. All this time i just blogging so someone could read my writings. this time is different. create this blog to just write what is in my mind. so. teras | teras, beranda, serambi, petak, anglong, tingkat in English is terrace. A terrace is an external, raised, open, flat area in either a landscape (such as a park or garden) near a building. so its the face in a building. sometimes terrace is already inside the building behind the facad. ...

23 November 2022 | prima adi