Fragments: October 21
LLM generated prose generates rejection, cancellation of Grade Hopper celebration, using AI to assist debugging unfamiliar code
Source feed
78 items
A website on building software effectively
LLM generated prose generates rejection, cancellation of Grade Hopper celebration, using AI to assist debugging unfamiliar code
When software meets hardware, engineering principles remain the same, but the defaults change—from testing devices to handling disconnection, energy, diagnostics, observability, and recovery.
Building a reusable slide template with colour palettes, typography, and simple layouts.
An experiment asks whether a behavioral harness can improve the reliability of cost-optimized coding models.
Topics: Programming
Entities: Programming
walking the same path as the Krell, Ola Bini banished, interviewed at DDD Europe with Eric Evans, future of reading code, latest Google model knows less, but pretends less
Entities: Google
When slides help a presentation, and the principles that make them work.
A practice we expect a team to do, unless there's a better one for their particular context
agents make programming harder, creating a breakaway agent needs super-persistence, AI needs a conscience more than consciousness, theses on AI regulation
Topics: Programming
Entities: Programming
the risk from AI is wiring it into everything, colorful syntax highlighting impedes comprehension, what Forward Deployed Engineer should mean
A guide to exchanging criticism or praise in teams
a short post
two worrying explanations around the RubyGems hack, agents aren't so much super-intelligent as super-persistent, Uncle Bob's harness is obviated, regulation of AI and competition with China