4 articles
One week testing Cline: the open-source VS Code AI agent that uses Claude API to write, edit, and debug code autonomously. Real user stories and honest trade-offs across three real projects.
Two weeks testing Windsurf, Codeium's AI code editor: Cascade agent mode, multi-file editing, real-time codebase understanding, and how it stacks up against Cursor and Copilot.
Two weeks of real project testing: tab completion, multi-file refactoring, codebase Q&A. Honest review with real user stories.
A week testing Replit Agent across five apps: todo list, dashboard, landing page, API, and polling app. Prompt-to-app in 30 seconds — but the edge cases tell a different story. Honest trade-offs from real use.
I tested Cody, Sourcegraph's AI coding assistant that indexes your entire codebase. Here's where it beats generic AI tools and where it still falls short.
I tested Continue.dev, the open-source AI coding assistant for VS Code and JetBrains. What it does better than Cursor and where it falls short.
Aider is an open-source AI coding assistant that works in your terminal. A deep dive into its git-aware editing, real results from developers, and honest trade-offs.
A week testing Replit Agent across five apps: todo list, dashboard, landing page, API, and polling app. Prompt-to-app in 30 seconds — but the edge cases tell a different story. Honest trade-offs from real use.
A week with Bolt.new: the AI app builder that turns prompts into working web apps. Real results from 5 projects and honest trade-offs from a skeptic.
Two weeks testing Codeium's free AI code completion across Python, React, and Rust projects. A Copilot subscriber confronts the free alternative and finds it surprisingly competitive — plus where it still can't keep up.
Two weeks testing Tabby, the open-source self-hosted AI code assistant. Free, private, and runs on your own GPU. Here's where it surprised a Copilot user — and where it still falls short.
Two weeks coding with Supermaven: the $10/month AI autocomplete that processes 1M tokens per second. Real results on a React app, a Python backend, and a Rust CLI tool. Honest trade-offs from real developers.
One week testing Cline: the open-source VS Code AI agent that uses Claude API to write, edit, and debug code autonomously. Real user stories and honest trade-offs across three real projects.
Two weeks testing Windsurf, Codeium's AI code editor: Cascade agent mode, multi-file editing, real-time codebase understanding, and how it stacks up against Cursor and Copilot.
Copilot's 2025 updates finally caught up to Cursor. Agent mode, multi-file editing, real codebase understanding. We tested them all head-to-head.
Two weeks of real project testing: tab completion, multi-file refactoring, codebase Q&A. Honest review with real user stories.