Skip to content
Open
Changes from all commits
Commits
File filter

Filter by extension

Filter by extension

Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
133 changes: 133 additions & 0 deletions artificial-intelligence/README.md
Original file line number Diff line number Diff line change
@@ -0,0 +1,133 @@
This guide exists to save time by showing which AI tools are being used and how, so everyone can learn from collective experience rather than starting from scratch.

Check failure on line 1 in artificial-intelligence/README.md

View workflow job for this annotation

GitHub Actions / lint

First line in a file should be a top-level heading

artificial-intelligence/README.md:1 MD041/first-line-heading/first-line-h1 First line in a file should be a top-level heading [Context: "This guide exists to save time..."] https://github.com/DavidAnson/markdownlint/blob/v0.37.4/doc/md041.md

AI is evolving quickly, and people at thoughtbot are experimenting with different tools in their work. Instead of everyone spending time evaluating tools on their own, this guide brings together our shared experiences so we can learn from each other and make faster, better-informed choices.

This makes the guide a living resource, updated regularly to reflect how our use of AI is changing. It is not a fixed set of rules, but a practical reference that saves time, helps people get started more quickly, and ensures our decisions are built on collective experience.

Here’s how it works:
• Survey and gather input: Once a quarter, we ask everyone what AI tools they are using, how they are using them, and what has or hasn’t worked.

Check failure on line 8 in artificial-intelligence/README.md

View workflow job for this annotation

GitHub Actions / lint

Hard tabs

artificial-intelligence/README.md:8:3 MD010/no-hard-tabs Hard tabs [Column: 3] https://github.com/DavidAnson/markdownlint/blob/v0.37.4/doc/md010.md

Check failure on line 8 in artificial-intelligence/README.md

View workflow job for this annotation

GitHub Actions / lint

Hard tabs

artificial-intelligence/README.md:8:1 MD010/no-hard-tabs Hard tabs [Column: 1] https://github.com/DavidAnson/markdownlint/blob/v0.37.4/doc/md010.md
• Create a shared guide: The survey responses are collected here to give visibility into what people are actually using. This lets everyone see the range of tools in play, learn from others’ experiences, and decide what they’d like to experiment with themselves.

Check failure on line 9 in artificial-intelligence/README.md

View workflow job for this annotation

GitHub Actions / lint

Hard tabs

artificial-intelligence/README.md:9:3 MD010/no-hard-tabs Hard tabs [Column: 3] https://github.com/DavidAnson/markdownlint/blob/v0.37.4/doc/md010.md

Check failure on line 9 in artificial-intelligence/README.md

View workflow job for this annotation

GitHub Actions / lint

Hard tabs

artificial-intelligence/README.md:9:1 MD010/no-hard-tabs Hard tabs [Column: 1] https://github.com/DavidAnson/markdownlint/blob/v0.37.4/doc/md010.md

# AI in Development Survey Q3 2025 – Summary

## 1. Frequency of AI Usage

Check failure on line 13 in artificial-intelligence/README.md

View workflow job for this annotation

GitHub Actions / lint

Headings should be surrounded by blank lines

artificial-intelligence/README.md:13 MD022/blanks-around-headings Headings should be surrounded by blank lines [Expected: 1; Actual: 0; Below] [Context: "## 1. Frequency of AI Usage"] https://github.com/DavidAnson/markdownlint/blob/v0.37.4/doc/md022.md
| Frequency | Count |

Check failure on line 14 in artificial-intelligence/README.md

View workflow job for this annotation

GitHub Actions / lint

Tables should be surrounded by blank lines

artificial-intelligence/README.md:14 MD058/blanks-around-tables Tables should be surrounded by blank lines [Context: "| Frequency | Count |"] https://github.com/DavidAnson/markdownlint/blob/v0.37.4/doc/md058.md
|-----------|-------|
| Daily | 12 |
| A few times a week | 2 |
| Occasionally | 3 |
| Rarely | 1 |
| Never | 1 |

> **Insight:** Most developers in the survey use AI daily in their work.
Copy link
Member

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Developers are the only people who responded to this survey?

---

## 2. AI Models Used

Check failure on line 26 in artificial-intelligence/README.md

View workflow job for this annotation

GitHub Actions / lint

Headings should be surrounded by blank lines

artificial-intelligence/README.md:26 MD022/blanks-around-headings Headings should be surrounded by blank lines [Expected: 1; Actual: 0; Below] [Context: "## 2. AI Models Used"] https://github.com/DavidAnson/markdownlint/blob/v0.37.4/doc/md022.md
| AI Model | Count |

Check failure on line 27 in artificial-intelligence/README.md

View workflow job for this annotation

GitHub Actions / lint

Tables should be surrounded by blank lines

artificial-intelligence/README.md:27 MD058/blanks-around-tables Tables should be surrounded by blank lines [Context: "| AI Model | Count |"] https://github.com/DavidAnson/markdownlint/blob/v0.37.4/doc/md058.md
|----------|-------|
| OpenAI (GPT-3, GPT-4, DALL-E) | 12 |
| Anthropic (Claude) | 11 |
| Google AI (Gemini, LaMDA, PaLM) | 2 |
| Open Source Models | 1 |
| CodeRabbit | 1 |

> **Insight:** OpenAI and Anthropic are the dominant AI models among respondents.
---

## 3. AI Tools / Software Used

Check failure on line 39 in artificial-intelligence/README.md

View workflow job for this annotation

GitHub Actions / lint

Headings should be surrounded by blank lines

artificial-intelligence/README.md:39 MD022/blanks-around-headings Headings should be surrounded by blank lines [Expected: 1; Actual: 0; Below] [Context: "## 3. AI Tools / Software Used"] https://github.com/DavidAnson/markdownlint/blob/v0.37.4/doc/md022.md
| Tool / Software | Count |
|----------------|-------|
| ChatGPT | 12 |
| GitHub Copilot | 7 |
| Claude Code | 6 |
| Cursor | 3 |
| Roo Code | 1 |
| gptel in Emacs | 1 |
| Other | 4 |

> **Insight:** ChatGPT is the most commonly used AI tool, followed by Copilot and Claude Code.
---

## 4. Where AI is Used
| Location | Count |
|---------|-------|
| Inside the code editor | 12 |
| In the browser | 8 |
| In the terminal | 5 |
| Native apps | 2 |
| ChatGPT software/app | 2 |
| Sometimes in desktop apps | 1 |

> **Insight:** Developers mostly use AI integrated into their code editor, with secondary use in browsers or terminals.
---

## 5. Tasks AI is Used For
| Task | Count |
|------|-------|
| Writing code | 14 |
| Debugging code | 11 |
| Writing tests | 8 |
| Explaining unfamiliar code | 11 |
| Planning features / architecture | 8 |
| Writing commit messages | 2 |
| Generating documentation | 3 |
| Reminders / recalling knowledge | 1 |
| Autocomplete | 1 |
| Finding best practices / options | 1 |

> **Insight:** Core programming tasks like writing code, debugging, and understanding code are the most common AI applications.
---

## 6. Prompting Strategies
| Prompt Strategy | Count |
|----------------|-------|
| No specific strategy | 5 |
| Zero-shot prompting | 4 |
| Few-shot prompting | 5 |
| Chain-of-thought prompting | 2 |
| Persona-based prompting | 3 |
| Constraint-based prompting | 5 |
| Iterative prompting | 8 |
| Asking clarifying questions | 7 |
| Using templates / pre-defined prompts | 2 |
| Giving hard constraints (e.g., tests must pass) | 1 |

> **Insight:** Iterative prompting and asking clarifying questions are highly used strategies. Many developers combine multiple strategies.
---

## 7. Agentic AI Usage
| Usage | Count |
|-------|-------|
| Yes, frequently | 3 |
| Yes, occasionally | 3 |
| No, not yet | 10 |
| Never | 1 |

**Agentic AI Tasks:**
- Automated refactoring across multiple files
- Generating end-to-end features from a high-level description
- Automated testing and bug fixing workflows
- Intelligent code migration/upgrades
- Research and exploration of new technologies

> **Insight:** Most developers are **not yet using agentic AI**, but those who do focus on automation and feature generation.
---

## 8. AI Tools by Task
| Task | Tools |
|------|-------|
| **Writing Code** | ChatGPT, Copilot, Claude, Cursor, Other |
| **Debugging** | ChatGPT, Copilot, Claude, Cursor, Other |
| **Writing Tests** | Copilot, Claude, Cursor, Other |
| **Code Review** | Copilot, Claude, Other |
| **Planning & Architecture** | ChatGPT, Claude, Other |
| **Researching** | ChatGPT, Cursor, Other |

> **Insight:** ChatGPT and Copilot dominate coding and debugging, while Claude is frequently used for planning, testing, and research.