How I locally fine-tuned Whisper using my own voice data and minimal effort
Off-the-shelf Whisper models are impressive. Whisper is able to translate and transcribe a wide variety of languages while using consumer hardware. One reason it’s so good came from being able to leverage way more semi-self-supervised data than models before it,...
-
Making Firefox's AI Chatbot local-only with Lemonade
Firefox has a built-in AI chatbot sidebar which lets you query LLMs in a simple, efficent UI while browsing the web. The cool thing is that it supports local inference...
-
LLM generated 3D modeling is the way to go
Using LLMs and AI to generate new, complex, but reusable models via OpenSCAD and scripting is truly amazing. My journey in 3D modelling is still early, but I started with...
-
Automate Everything: LLMs and Bash Scripts Mean Saved Time
The landscape of software development is constantly evolving, and with the advent of sophisticated Large Language Models (LLMs), we’re seeing new possibilities emerge for automating tasks. However, relying on AI...
-
Why do Canadian News Sites Track Me so Much?
I noticed while reading the news on the CBC (Canadian Broadcasting Corporation) that ublock origin blocked 11 domains from making a connection which struck me as odd. Why was Canada’s...
-
Rethinking Library Dependencies in the GenAI age
The rise of Generative AI (GenAI) is transforming how developers approach common development challenges. One emerging trend is the ability to use GenAI to generate custom, lightweight solutions that reduce...
-
Using hCaptcha instead of Google's ReCaptcha with Vue and Express
For much the same reasons as my last captcha related post, when needing a form of bot deterrence, it makes sense to use a captcha provider. At fixed.link (a now...
-
Using Google's Captcha V3 with Express and Node
Before getting started, check out the newer version of this article where I use hCaptcha with express/Vue. hCaptcha is a non-google dependent captcha service. Otherwise, enjoy.
-
Considerations when supporting a legacy software stack
Another year and another post about managing legacy web applications.
-
Build a URL Shortener with Node + Mongo
A while back, long before tonight, I built a Node.js application called Lyt.Haus that shortened URLs. Conceptually, it’s simple and not exactly groundbreaking—there are hundreds, if not thousands, of sites...
-
Building a Flask Application: Where to Start
Flask is a Python framework designed to provide a lightweight and highly flexible web server. Flask sells itself by stating how unopinionated it is, how few dependancies are required and...
-
Refreshing an old stack's front end
It’s obvious that refactoring or restructuring code too often is bad, and it’s especially obvious that completely switching an application’s core technology isn’t ideal either - but sometimes it is...
-
The Good and Bad After 12 Months with a System76 Lemur Pro 9
12 months ago, I bought a rather expensive ($1,600 Canadian) laptop from System76 - a vendor that focuses on Linux hardware and software. That being said, S76 did not manufacture...
-
Building a Startup on $10 a Month
When building Fixed.link, I prioritized minimizing spend and built my startup which served many thousands of users while spending only $10 a month. Early on in your startup journey, it’s...