You Don’t Need the Best Model. You Need a Better Loop.
There is a common assumption in AI-assisted software development: if the result is not good enough, you need a stronger model. My latest experiment suggests something different.
A Dying Star, a Swarm of Machine Birds and One Last Shield: The Making of PHØENIX: NEON RISING
Some games you play once and forget. Others stay with you for forty years. For me, it was the bird shooters of the early arcade era: formations sweeping in from the top of the screen, birds peeling off …
Read the article
4 min read
Welcome to Phooky.COM
Look up at the stars and not down at your feet. Try to make sense of what you see, and wonder about what makes the universe exist.
Be curious.
Selected articles
Claude Code Loops: Automation Is Not Intelligence
Claude Code loops can reduce micromanagement by allowing an AI agent to implement, test, debug, and repeat autonomously. But repetition alone does not produce better software. Without reliable verification, strict boundaries, and explicit stopping conditions, loops primarily automate assumptions—including incorrect ones.
AI Meets Programming Prowess: Building a Pong Game in Flutter
This article explores the synergy between GPT-4 and human expertise by taking on the challenge of creating a PONG game with Flutter in under 30 minutes.
More articles
Read more about Python programming, cloud computing and many more topics I like.
-
Dockerize Wagtail & PostgreSQL as a development environment
If you want to develop with Wagtail and PostgreSQL in a platform-independent way, using Docker is a good choice. In the following tutorial, I would like to share my approach.
-
Twitter Mining with Tweepy and mongoDB
If you want to analyze data on current topics and opinions, Twitter is often a very good data source. To save tweets from the Twitter Streaming API to a mongoDB database easily, I wrote this little Python script.
-
Twitter Sentiment Analysis with Azure's Text Analytics API
Creating a sentiment analysis with Twitter messages is not really easy due to the brevity of the texts and the frequently used irony and sarcasm. With the Azure Text Analytics API this works surprisingly well anyway.
