You can just ask AI to do things

kept by eddie oct 30

The key to effective AI use is agency: proactively asking AI agents to automate tedious development tasks like running servers and monitoring logs.

You have to beat the models at something
Engineers should leverage deep codebase familiarity and clear technical communication to stay valuable, as LLMs lack contextual understanding and struggle with human-like writing.
also on Sean Goedecke, #ai-coding, #productivity, #developer-tools
Why I Ignore The Spotlight as a Staff Engineer - Lalit Maganti
A senior staff engineer at Google argues that long-term stewardship of infrastructure systems yields more impact than chasing high-visibility product work.
also on Sean Goedecke, #developer-tools
LLMs as Judges: Measuring Bias, Hinting Effects, and Tier Preferences
LLMs used as judges to evaluate other LLMs exhibit measurable self-preference bias, with GPT showing the strongest self-bias and Claude the weakest, while revealing model identities through hinting changes judge behavior unevenly across vendors and domains.
also on GPT, Gemini, Claude, #ai-coding, #developer-tools
Flash Papers — academic papers, reimagined
Flash Papers converts any arXiv or ChinaRxiv paper into an interactive, AI‑powered explainer site with charts and diagrams you can manipulate.
also on GPT, Gemini, Claude, #developer-tools
AI coding assistants are getting worse?
The article argues that AI coding assistants are degrading because training data quality drops as inexperienced coders accept flawed code, creating a feedback loop.
also on GitHub Copilot, Gemini, Claude, #ai-coding, #developer-tools