Urgent.News

What's breaking now, across thousands of outlets.

AI

Everything, All at Once

By the end of this post, you'll learn how I work faster with coding agents and run more work in parallel. Development looks very different form a year ago. I can't remember the last time I wrote a line of code myself. Over time, I've built workflows and systems that let me trust agents with much more of the development process. I was also fortunate to start working on a large greenfield project…

I've been using coding agents to speed up my development process and run multiple tasks simultaneously. Development has changed a lot in the past year, and I no longer write code myself. Instead, I've developed workflows and systems that allow me to trust agents with more of the development process. When working on a single feature, like building a login feature, I ask my agent to write a technical specification, breaking it down into verifiable phases, and tagging phases that can be worked on in parallel.

I store the spec and plan in a temporary directory in my codebase for later reference. This approach has increased my productivity and confidence in the agent's work. When working on multiple features simultaneously, I use Git worktrees to create separate branches checked out and worked on at the same time, allowing multiple agents to work on different features independently.

I create a worktree for each feature, implement the plan, and run subagents for parallel work. This method has improved my workflow and allowed me to ship more code faster.

Written by urgent.news from Dev.to's reporting — not their text. Machine-written — may contain errors; check the original before relying on it.

Read the original at dev.to →

More in AI

Debugging Is the Killer App for Free Model Tokens — Here's the Workflow

Most developers treat free model tokens as a code generation budget. They ask for snippets, refactors, and explanations, then wonder why the tokens disappear without making their codebase measurably…

  • Debugging is pattern-matching task, not code generation
  • Collect recent logs, extract error block and context
  • Python script automates workflow with free model API

Cooper Sharp Proves That American Cheese Can Be Great Cheese

John Moltz, in his weekly members-only column for Six Colors (no gift links, but Moltz’s column is worth the subscription in and of itself): This week Anthropic announced that in order to identify AI…

  • Anthropic plans to watermark AI-generated text
  • Tech commentator John Gruber criticizes strategy
  • Concerns about model training and AI inbreeding

I Built an AI That Fixes Terraform Drift Automatically

I Built an AI That Fixes Terraform Drift Automatically If you've worked with Terraform long enough, you know the feeling. You run terraform plan and suddenly there are 12 unexpected changes.

  • tfdrift remediate AI tool fixes Terraform drift automatically
  • Detects drift across all Terraform workspaces and classifies severity
  • Uses Claude or GPT-4o to generate remediation file for human review

More from Sunday 23 August →