Jupyter is a useful environment for data exploration. Despite the many excellent visualization...

A goal for 2026 is to camp more. I bought a National Parks Pass in December with the idea of extended...

Microsoft released markitdown, a tool for converting files to Markdown such as PDFs, PowerPoint, Word, and Excel. In addition, it can perform audio transcription and image description and convert standard formats like HTML, CSV, JSON, and XML.

Maximum effort! Not really. We're building an agent to query the database for this installment. We'll...

In the previous post, we created documents based on the tables in the Northwind Trader database. We... Tagged with nebius, text2sql, llm, generativeai.

Nebius AI Studio is an inference service offering state-of-the-art open-source large language models.... Tagged with nebius, rag, text2sql, llm.

Building a Data Pipeline with DLT In this example, a company provides IoT data through a...

Jigsaw Stack's API provides various services, such as web scraping and LLM prompting. I tried out the...

This is a guest post on the Continue Blog by Sophia Parafina, a developer advocate who has previously worked at Pulumi, Anaconda, and Docker. Continue is an open-source AI code assistant. It's a compound AI system that uses a set of models (chat,

Retrieval augmented generation (RAG) is a popular way to use current or proprietary data with Large Language Models (LLMs). There are many articles describing how to perform RAG. Typically, they involve encoding data as vectors and storing the…

In a previous post, I demonstrated how to implement RAG using the Load-Transform-Embed-Store workflow. I've updated the RAG Step-by-Step repository with an open-source example. Why open source? The most common reason for using open source versus a…

In the previous post, we went through the steps to build a backend for an RAG application. This is cool, but not being able to demonstrate it is no fun. The process for RAG is to retrieve data from the vector database and insert it into a prompt. To…

Frank Denneman of VMWare recently published RAG Architecture Deep Dive which defines the process for RAG applications. The article defines a Load-Transform-Embed-Store workflow. Let's implement the workflow in practice. For this tutorial, I'm using…

It's been fun playing with LLMs with a CPU. However, the novelty wears off as I watch the completion slowly display word by word. Enter the GPU, I have an older Ubuntu gaming laptop with a GPU I purchased for machine learning (I haven't played a game…

Getting llamafile running with a GPU can be a little tricky. For reference, I have a four-year-old Kubuntu laptop with a Nvidia RTX 2060. Check to see if the correct Nvidia driver is installed. nvida-smi If the driver is installed, you will see a…

Llamafile is a Mozilla project that runs open source Large Language Models such as Llama-2-7B, Mistral 7B, or any other models in the GGUF format. Containerize Llamafile The Dockerfile builds, containerizes llamafile, and runs it in server mode. It…

Local AI is the ability to run artificial intelligence models on commodity hardware. Many models can run on CPUs without a GPU, though a GPU is an advantage when running models. This means that anyone with a relatively recent computer build or a…

If you are installing Hugging Face transformers on a M2 macOS machine, you might run into this error × Getting requirements to build wheel did not run successfully. │ exit code: 1 ╰─> [5 lines of output] Package sentencepiece was not found in the…

There are situations where you need access to AWS resources that are not in Lightsail, such as files on an EC2 server or the need to connect to an AWS RDS such as PostgreSQL or MariaDB. It is possible to use these resources via VPC (Virtual Private…