「Cursor(カーソル)」と「Dify(ディフィ)」
最近、調べ物や翻訳、議事録作成などでAIを使うのがすっかり日常になってきた。もう「第3の社員」と言ってもいいくらい、頼る時間が増えている。
ただ、周りのAI活用の猛者たちの話を聞くと、自分はまだまだ初級者だと感じる。彼らが口を揃えて勧めるのが「Cursor(カーソル)」と「Dify(ディフィ)」というツールだ。
カーソルはAIがプログラミングコードを自動生成してくれるエディタ、ディフィは複数のAIを組み合わせてワークフローを作れるツール。まさに“AI中上級者”の必須アイテムらしい。
そんなとき、ある人のブログを読んだ。泊まり込みで「AI学校(研修ワークショップ)」に参加した体験記だった。
参加者はプログラミング経験のない経営者やビジネスパーソンたち。iPhoneの録音アプリに音声で自分のアイデアを吹き込み、それをChatGPTに整理させ、カーソルでコード化。30分もかからずに見事なWebサイトが完成したという。さらにディフィを使って問い合わせフォームや自動応答まで実装していったと書かれていた。
そのプロセスを読んで、素直に「すごいな」と思った。もうITシステム開発のやり方自体が変わってきているのだろう。すでに工数が従来の1/3〜1/5になったという事例もあるらしい。
記事を読みながら「これは自分も学んでみるべきかもしれない」と考えた。AIを便利な道具として使うだけじゃなく、仕組みとして理解し、仕事に組み込んでいく段階に入る時期なのかもしれない。
ーーーー
ーーーー
“Cursor” and “Dify”
These days, using AI for research, translation, or meeting notes has become part of my daily routine. It’s reached the point where I sometimes call it my “third employee,” given how much I rely on it.
Still, when I hear stories from people around me who are true power users of AI, I realize I’m probably still at the beginner level. They often mention two tools in particular: Cursor and Dify.
Cursor is an editor where AI automatically generates programming code, and Dify is a platform that lets you combine multiple AIs to build workflows. They seem to be essential tools for anyone who wants to call themselves an “intermediate to advanced AI user.”
Recently, I came across a blog post written by someone who attended an overnight “AI school” workshop.
The participants weren’t engineers—they were managers and entrepreneurs with no programming background. They started by speaking their website ideas into an iPhone recording app, then had ChatGPT organize the messy transcription into clear requirements. Those requirements were fed into Cursor, which produced the code. Within just 30 minutes, a fully functioning website was up and running. From there, they used Dify to add a contact form and even an automated response system.
Reading about this process, I couldn’t help but think, “That’s incredible.” It’s clear that the way IT systems are developed is already changing. There are even cases where project workloads have been cut to one-third or one-fifth of what they used to be thanks to AI-driven coding.
As I read the story, I thought: Maybe I should try learning this too. It feels like the time has come to go beyond just using AI as a convenient tool, and instead understand its mechanisms—so I can truly integrate it into my work.
カテゴリー
– Archives –
– other post –
– Will go to Mars Olympus –
– next journey Olympus on Mars through Space Travel –
– 自己紹介 インタビュー –
– Books –
*“Yesterday, I Went to Mars ♡”*
- Age 46, a Turning Point: What the Inventor of Morse Code Taught Me 2025年9月20日This episode looks at the life of Samuel Morse, the inv […]
- The Power of Steady Effort 2025年9月19日For 17 years, Makoto Optical has continued serving cust […]
- "The Day the Turquoise Disappeared - The Power of Meaning in Jewelry" 2025年9月19日A reflection on a piece of Indian jewelry whose turquoi […]
- The Physical Installation of Skill 2025年9月18日On today's episode, we discuss a fascinating new techno […]
- Stability Is a Fortunate Exception 2025年9月16日This episode reflects on the idea that stability is nev […]