Appears on
Articles21
Benchmark Comparison: Qwen3.6-35B-A3B Outperforms Claude Opus 4.7 in Pelican Image Generation Test
Anthropic Restricts Claude Mythos AI Access Through Project Glasswing for Security Research
Analyzing Hacker News User Behavior Through Comment History Profiling
The article describes a method for profiling Hacker News users by analyzing their last 1,000 comments using the Algolia Hacker News API. The author presents a 'mildly dystopian' prompt of 'Profile this user' and explains how to access comment data through the API, which has open CORS headers allowing direct browser access. The piece explores the technical im
Key Researcher Departs Alibaba's Qwen AI Team Amid Concerns About Project's Future
The article discusses recent high-profile departures from Alibaba's Qwen AI research team, particularly the resignation of lead researcher Junyang Lin, who was instrumental in developing the Qwen 3.5 family of open weight models. The author expresses concern that these departures might signal trouble for the Qwen project, potentially making Qwen 3.5 the team
Showboat and Rodney: New Tools for Demonstrating and Testing Coding Agent Output
StrongDM Implements Fully Automated AI Software Factory Without Human Code Review
ChatGPT's Code Interpreter Gains Major Upgrades: Bash Commands, Package Installation, and File Downloads
Analyzing Daft Punk's "Harder, Better, Faster, Stronger" Tempo: Is 123 BPM an Intentional Musical Joke?
2025 LLM Industry Review: Reasoning Models, Open Source Advances, and AI Safety Debates
Professional Responsibility in Software Development: Delivering Proven Code in the AI Era
The article criticizes the practice of junior software engineers using AI coding assistants to generate large amounts of untested code and expecting code reviews to handle quality assurance. The author argues this is disrespectful to colleagues and a dereliction of professional duty, emphasizing that software engineers' core responsibility is to deliver prov
Patterns and Insights for Building Single-File HTML Applications
OpenAI Implements Skills Support in ChatGPT and Codex CLI Following Anthropic's Model
Developer Reverse Engineers OpenAI's Codex CLI to Access GPT-5-Codex-Mini Model Directly
New Research Papers Address LLM Security and Prompt Injection Vulnerabilities
Coding Like a Surgeon: Using AI Tools to Focus on High-Value Design Work
geoffreylitt.com7mo ago
Running DeepSeek-OCR on NVIDIA Spark Hardware Using Claude Code Automation
GPT-5's search capabilities prove surprisingly effective, changing the chatbot-as-search-engine calculus
Understanding 'Vibe Coding' and the Challenges of Legacy Code
blog.val.town10mo ago
Testing the GLM-4.5 Air Model: Writing Space Invaders in JavaScript on an Older Laptop
