Coding
11 articles about coding.
AI Autocomplete Is Sufficient 90% of the Time - When You Need More
AI autocomplete handles most coding tasks. But when do you actually need a full agent-assisted development workflow? It depends on what you're building.
Letting AI Coding Agents Use Real Debuggers Instead of Guessing
AI coding agents guess at bugs by reading code. Giving them access to real debuggers - breakpoints, stack traces, variable inspection - makes them
AI Really Killed Programming for Me
AI did not kill programming - it shifted the competitive advantage from writing code to understanding systems. The skill that matters now is knowing what to
I Forgot How to Code After Using AI Agents
Anthropic research confirms it: AI coding assistance reduces skill formation by 17%. Here's what atrophies, what grows, and how to stay sharp while using AI tools heavily.
GPT 5.4 vs Opus 4.6: Simplicity vs Over-Architecture
Opus 4.6 picks the simplest approach that works. GPT 5.4 tends to over-architect solutions. For desktop agent development, simplicity wins.
How Are In-Office Dev Jobs Now? Coding Time Dropped to 30%
In-office developer roles have shifted dramatically. Actual coding is now about 30% of the job - the rest is reviewing AI output, writing specs, and
Is Cursor Falling Behind Claude Code?
Claude Code reads, edits, runs, and tests in one loop. Cursor still separates these steps. The integrated loop is winning for developers who want to ship
M4 Pro with 48GB Memory for Local Coding Models?
48GB of unified memory on an M4 Pro fits 70B parameter models at Q4 quantization. Local inference for privacy-sensitive work and overnight batch processing.
Agent Orchestrators vs Parallel Sessions with Worktrees
Comparing agent orchestration patterns vs parallel sessions with git worktrees. Real isolation wins for coding tasks because each agent gets its own workspace.
GPT's Lazy File Patching Problem - Partial Copies and Broken Imports That Waste Your Time
GPT's auto mode picks the stronger model for complex tasks, but its file patching is infuriating. Partial copies leave broken imports and missing code.
Running 5 AI Coding Agents in Parallel - Setup, Coordination, and Real Tradeoffs
How to run multiple Claude Code agents simultaneously in a terminal IDE, how to manage context sharing between them, and what the practical ceiling actually is.