#ai
Reviewing OpenAI's Deep Research: A good starting point
February 04, 2025Deep Research is an interesting first implementation that can provide highly engaging outputs. However, it's not yet anything near what an analyst would produce, especially due to the suboptimal research process.
Accessing OpenAI's Operator outside the USA
January 29, 2025A simple approach to accessing OpenAI's Operator and Sora outside the USA, and especially in Europe.
OpenAI Operator review: Currently too limited but reasons to be hopeful
January 29, 2025OpenAI's Operator is an interesting product/research preview. It has a lot of potential and it's only a matter of time before it provides significant value to its users. But for now, its reasoning and browser skills are too limited.
Enriching or modifying an Anki deck with LLMs
January 03, 2025If you want to make large-scale changes to an Anki deck, OpenAI's GPT or Perplexity are great options. As there are no native integrations yet with those tools, I share a small script to get the same result.
Using OpenAI's o1 models for code reviews (and my custom rake task)
December 18, 2024With a simple rake task, you can generate a prompt to have o1 provide a comprehensive review of your code.
Trying out the paid Bug Finder in Cursor
December 12, 2024Cursor has an experimental Bug Finder that's quite expensive and is paid for at every run. I was curious to try it out and share in case anyone's curious.
Using embeddings to display related articles with Supabase in Ruby on Rails
October 31, 2024A short tutorial on how to use Supabase to store embeddings on blog posts to display related articles.