Tag: ai-coding
Page 14 of 48 • 521 posts total
-
Local LLMs with Ollama: Picking a local model for the specific task, not for the benchmark
A field-tested take on picking a local model for the specific task, not for the benchmark with Local LLMs with Ollama: what it rewards, where it breaks, and how to keep the workflow honest.
-
Agentic coding: Budgets as the only feature that stops a bad loop
A field-tested take on budgets as the only feature that stops a bad loop with Agentic coding: what it rewards, where it breaks, and how to keep the workflow honest.
-
MCP servers: Smallest MCP server that earns its keep
A field-tested take on the smallest MCP server that earns its keep with MCP servers: what it rewards, where it breaks, and how to keep the workflow honest.
-
AI coding workflows: Review cycle that keeps agent output honest
A field-tested take on the review cycle that keeps agent output honest with AI coding workflows: what it rewards, where it breaks, and how to keep the workflow honest.
-
Prompt engineering for code: Shortest prompt that still passes review
A field-tested take on the shortest prompt that still passes review with Prompt engineering for code: what it rewards, where it breaks, and how to keep the workflow honest.
-
Cline and Roo Code: Picking a model per task instead of per editor
A field-tested take on picking a model per task instead of per editor with Cline and Roo Code: what it rewards, where it breaks, and how to keep the workflow honest.
-
JetBrains AI Assistant: Letting Junie drive while the IDE still owns correctness
A field-tested take on letting Junie drive while the IDE still owns correctness with JetBrains AI Assistant: what it rewards, where it breaks, and how to keep the workflow honest.
-
Windsurf: Cascade as a deliberate planning surface, not an auto-pilot
A field-tested take on Cascade as a deliberate planning surface, not an auto-pilot with Windsurf: what it rewards, where it breaks, and how to keep the workflow honest.
-
Cursor: Writing .cursorrules that actually stay short and enforceable
A field-tested take on writing .cursorrules that actually stay short and enforceable with Cursor: what it rewards, where it breaks, and how to keep the workflow honest.
-
Google Gemini: Feeding an entire service plus its tests into a single prompt
A field-tested take on feeding an entire service plus its tests into a single prompt with Google Gemini: what it rewards, where it breaks, and how to keep the workflow honest.
-
ChatGPT Codex: Letting Codex open the PR while you stay the reviewer of record
A field-tested take on letting Codex open the PR while you stay the reviewer of record with ChatGPT Codex: what it rewards, where it breaks, and how to keep the workflow honest.