April 2025
30 published posts filed under April 2025. Each article shows its actual publication date.
-
ChatGPT Codex: First time we stopped Codex mid-task and why
A field-tested take on the first time we stopped Codex mid-task and why with ChatGPT Codex: what it rewards, where it breaks, and how to keep the workflow honest.
-
Claude Code: When headless Claude Code became part of our CI
A field-tested take on when headless Claude Code became part of our CI with Claude Code: what it rewards, where it breaks, and how to keep the workflow honest.
-
GitHub Copilot: Measuring Copilot adoption without lying to ourselves
A field-tested take on measuring Copilot adoption without lying to ourselves with GitHub Copilot: what it rewards, where it breaks, and how to keep the workflow honest.
-
Local LLMs with Ollama: Offline code reviewer that survived a flaky Wi-Fi week
A field-tested take on an offline code reviewer that survived a flaky Wi-Fi week with Local LLMs with Ollama: what it rewards, where it breaks, and how to keep the workflow honest.
-
Agentic coding: Long-running agent that finally paused itself
A field-tested take on the long-running agent that finally paused itself with Agentic coding: what it rewards, where it breaks, and how to keep the workflow honest.
-
MCP servers: Observability gaps that bit us exactly once
A field-tested take on observability gaps that bit us exactly once with MCP servers: what it rewards, where it breaks, and how to keep the workflow honest.
-
AI coding workflows: Acceptance tests first, agents second, as a beginner mantra
A field-tested take on acceptance tests first, agents second, as a beginner mantra with AI coding workflows: what it rewards, where it breaks, and how to keep the workflow honest.
-
Prompt engineering for code: Style guides for both humans and models, a beginner attempt
A field-tested take on style guides for both humans and models, a beginner attempt with Prompt engineering for code: what it rewards, where it breaks, and how to keep the workflow honest.
-
Cline and Roo Code: Running Cline against a local LLM for the first time
A field-tested take on running Cline against a local LLM for the first time with Cline and Roo Code: what it rewards, where it breaks, and how to keep the workflow honest.
-
JetBrains AI Assistant: When Junie finished the task we had been avoiding
A field-tested take on when Junie finished the task we had been avoiding with JetBrains AI Assistant: what it rewards, where it breaks, and how to keep the workflow honest.
-
Windsurf: Beginner comparison: Windsurf, Cursor, VS Code Agent
A field-tested take on a beginner comparison: Windsurf, Cursor, VS Code Agent with Windsurf: what it rewards, where it breaks, and how to keep the workflow honest.
-
Cursor: MCP inside Cursor as an early glimpse of the toolbelt
A field-tested take on MCP inside Cursor as an early glimpse of the toolbelt with Cursor: what it rewards, where it breaks, and how to keep the workflow honest.
-
Google Gemini: Writing Mermaid diagrams Gemini could then critique
A field-tested take on writing Mermaid diagrams Gemini could then critique with Google Gemini: what it rewards, where it breaks, and how to keep the workflow honest.
-
ChatGPT Codex: Budgeting Codex tasks when usage was still a black box
A field-tested take on budgeting Codex tasks when usage was still a black box with ChatGPT Codex: what it rewards, where it breaks, and how to keep the workflow honest.
-
Claude Code: Turning a chaotic commit history into a tidy narrative
A field-tested take on turning a chaotic commit history into a tidy narrative with Claude Code: what it rewards, where it breaks, and how to keep the workflow honest.
-
GitHub Copilot: Premium request debate, before anyone had real data
A field-tested take on the premium request debate, before anyone had real data with GitHub Copilot: what it rewards, where it breaks, and how to keep the workflow honest.
-
Local LLMs with Ollama: Caching embeddings locally for a beginner RAG
A field-tested take on caching embeddings locally for a beginner RAG with Local LLMs with Ollama: what it rewards, where it breaks, and how to keep the workflow honest.
-
Agentic coding: How a junior engineer earned their first autonomous task
A field-tested take on how a junior engineer earned their first autonomous task with Agentic coding: what it rewards, where it breaks, and how to keep the workflow honest.
-
MCP servers: Versioning MCP servers like APIs, a first-year habit
A field-tested take on versioning MCP servers like APIs, a first-year habit with MCP servers: what it rewards, where it breaks, and how to keep the workflow honest.
-
AI coding workflows: Choosing boredom over brilliance in the first year
A field-tested take on choosing boredom over brilliance in the first year with AI coding workflows: what it rewards, where it breaks, and how to keep the workflow honest.
-
Prompt engineering for code: When a prompt quietly became a feature spec
A field-tested take on when a prompt quietly became a feature spec with Prompt engineering for code: what it rewards, where it breaks, and how to keep the workflow honest.
-
Cline and Roo Code: Memory banks as a beginner experiment in agent continuity
A field-tested take on memory banks as a beginner experiment in agent continuity with Cline and Roo Code: what it rewards, where it breaks, and how to keep the workflow honest.
-
JetBrains AI Assistant: Pairing Code With Me with AI Assistant for remote review
A field-tested take on pairing Code With Me with AI Assistant for remote review with JetBrains AI Assistant: what it rewards, where it breaks, and how to keep the workflow honest.
-
Windsurf: Pausing Cascade versus pausing yourself
A field-tested take on pausing Cascade versus pausing yourself with Windsurf: what it rewards, where it breaks, and how to keep the workflow honest.
-
Cursor: When agent mode decided to write tests we never wrote
A field-tested take on when agent mode decided to write tests we never wrote with Cursor: what it rewards, where it breaks, and how to keep the workflow honest.
-
Google Gemini: Verbosity we trimmed in month two
A field-tested take on the verbosity we trimmed in month two with Google Gemini: what it rewards, where it breaks, and how to keep the workflow honest.
-
ChatGPT Codex: Diff-first review habit we learned from Codex
A field-tested take on the diff-first review habit we learned from Codex with ChatGPT Codex: what it rewards, where it breaks, and how to keep the workflow honest.
-
Claude Code: First time Claude Code said no to a bad instruction
A field-tested take on the first time Claude Code said no to a bad instruction with Claude Code: what it rewards, where it breaks, and how to keep the workflow honest.
-
GitHub Copilot: Discovering Copilot Edits and rethinking the pull request
A field-tested take on discovering Copilot Edits and rethinking the pull request with GitHub Copilot: what it rewards, where it breaks, and how to keep the workflow honest.
-
Local LLMs with Ollama: Moment a local 7B beat the cloud on a real task
A field-tested take on the moment a local 7B beat the cloud on a real task with Local LLMs with Ollama: what it rewards, where it breaks, and how to keep the workflow honest.