Blog on www.benmarte.com

How Are You Using AI to Improve Your Workflow?

Mon, 16 Feb 2026 00:00:00 +0000

I’ll be honest, for the longest time I was using AI the same way everyone else was. Open a chat, paste some code, ask a question, copy the answer, move on. It worked, but it wasn’t exactly life-changing. I was basically using a Ferrari to go grocery shopping. 🙃

Then I discovered Claude Code and everything changed. Not overnight though, it took me months of trial and error, breaking things, getting frustrated, and slowly figuring out how to make it actually work the way I needed it to. What I ended up with is a setup that honestly feels less like a tool and more like having a team of developers sitting next to me. Let me walk you through how I got here.

How to run your own private AI assistant locally

Wed, 13 Mar 2024 00:00:00 +0000

In this article we will discuss how to run your own AI assistant using Ollama as the LLM (Large Language Model) and Chatbox in order to have a desktop GUI (Graphical User Interface) for your assistant.

Prerequisites

Install Ollama
Install Chatbox

Selecting which model to use

Once you have installed Ollama you need to select which model you want to use, there are many different models you can choose from for different types of tasks so feel free to explore the different models and select the one that suits your needs.

HMR in k8s with Tilt

Sun, 13 Aug 2023 00:00:00 +0000

Kubernetes is a great tool to deploy your applications quickly and reliably. However, no one seems to mention that it can break your development flow, especially if you use a framework like react or nextjs.

Challenges like runtime environment variables not being properly passed into your containers, covered in this previous article, or the constant need to build, load, and set new pod images with every change to your code, which is what we will be discussing in this article.

NextJS in Docker

Sat, 25 Jun 2022 00:00:00 +0000

Being a Full Stack R&D Developer at Raft comes with a few interesting challenges to solve. In this blog post, we will discuss one of the latest problems my team and I ran into during the development of DataFabric.

DataFabric, is a pretty complex app that has integrations with many other apps (Datahub, Superset, Kafka, Trino to name a few) and cutting-edge technologies (Docker, Kubernetes, NextJS, and more). Integrating with these technologies and making them play nice with each other can sometimes lead to going down a rabbit hole that has no end in sight.

Hugo, S3, Cloudfront and Route 53, Oh My!

Sun, 06 Feb 2022 00:00:00 +0000

After spending more time than necessary to get a Hugo site working on AWS with Cloudfront I decided to write this blog post not just for the other poor souls running into this problem but also for myself because I tend to forget shit 🙃

Anyways, let’s cut to the chase it’s a Sunday and I should be getting drunk and the edible hasn’t kicked in yet 😏

The Problem

Hosting a static site on S3 is pretty straight forward. Upload your HTML files to your bucket, make the bucket public, and enable static website hosting. That is until you need to use a custom domain with an SSL certificate for your site, you’re already invested in the AWS ecosystem so you say I’ll just use cloudfront 🥴 and so the pain begins.

Open EMR Backup and Restore Docker Procedure

Thu, 04 Mar 2021 00:00:00 +0000

One of the most important things in development is disaster recovery, using Open EMR and being HIPAA compliant you must have a solid understanding on how to backup/restore your Open EMR instance because like Murphy says: “Anything that can go wrong will go wrong”. With that in mind I am going to show you today how-to backup/restore your Open EMR instance, if you use an Open EMR instance from the AWS Marketplace then this is for you.

How to use Open EMR v6 API Password Grant Flow

Tue, 26 Jan 2021 00:00:00 +0000

Pre-Requisites

Open EMR v6 installed
Postman or an alternative tool/app to test API calls

After I upgraded my Open EMR instance without reading the changelog (never do this) and breaking my site (this is why) due to massive changes to the way their API works I had to then spend 4 days wrapping my brain around how the new API oauth2 flows work with Open EMR v6. Luckily my app is a POC and nothing to be concerned about it being broken so I had time to work on it, the following is what I learned in my 4 days trying to get Open EMR v6 API working with my React app using the Password Grant Flow.