Skip to content

Commit d277b1f

Browse files
committed
Format extend-document-automation talk to match project documentation standards
1 parent f3da802 commit d277b1f

File tree

9 files changed

+443
-20
lines changed

9 files changed

+443
-20
lines changed
Lines changed: 35 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -0,0 +1,35 @@
1+
# Documentation Writing Style Guide
2+
3+
When writing or editing documentation in the `docs/` directory, follow these style guidelines:
4+
5+
## Writing Standards
6+
- **Reading level**: Write at 9th-grade reading level for accessibility
7+
- **No emojis**: Keep documentation professional and emoji-free
8+
- **Conversational tone**: Use "you", "we", "I" to make content personal and engaging
9+
- **Active voice**: Prefer active over passive voice
10+
- **Specific examples**: Include concrete examples and metrics when possible
11+
12+
## Technical Content
13+
- **Actionable insights**: Focus on practical, implementable advice over theory
14+
- **Code examples**: Include working code snippets with proper syntax highlighting
15+
- **Performance metrics**: Include specific numbers and improvements when available
16+
- **Company attribution**: Reference companies and speakers for credibility
17+
18+
## Markdown Formatting
19+
- **Headers**: Use proper hierarchy (H1 for titles, H2 for main sections, H3 for subsections)
20+
- **Bold text**: Use `**bold**` for emphasis and labels
21+
- **Code blocks**: Use triple backticks with language specification
22+
- **Lists**: Use `-` for bullet points with consistent formatting
23+
- **Links**: Use descriptive link text, not "click here"
24+
25+
## Content Organization
26+
- **Clear structure**: Organize content with logical flow and clear headings
27+
- **Key takeaways**: Highlight main insights with `**Key Takeaway:**` format
28+
- **Scannable content**: Use bullet points, short paragraphs, and clear headings
29+
- **Consistent formatting**: Match existing document styles in the same directory
30+
31+
## Technical Accuracy
32+
- **Verify examples**: Ensure all code examples work and are up-to-date
33+
- **Check links**: Verify all internal and external links are functional
34+
- **Consistent terminology**: Use the same terms throughout related documents
35+
- **Proper citations**: Reference sources and speakers appropriately
Lines changed: 39 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -0,0 +1,39 @@
1+
# MkDocs Navigation Standards
2+
3+
When updating the navigation in [mkdocs.yml](mdc:mkdocs.yml), follow these standards:
4+
5+
## Talk Navigation Format
6+
For talks in the navigation, use the format: `"Topic - Speaker Name (Company)": talks/filename.md`
7+
8+
### Examples:
9+
```yaml
10+
- "Coding Agents - Nik Pash (Cline)": talks/rag-is-dead-cline-nik.md
11+
- "RAG Anti-patterns - Skylar Payne": talks/rag-antipatterns-skylar-payne.md
12+
- "Semantic Search - Will Bryk (Exa)": talks/semantic-search-exa-will-bryk.md
13+
```
14+
15+
## Navigation Organization
16+
Organize talks into logical chapters and subsections:
17+
18+
### Chapter 5 Structure:
19+
```yaml
20+
- "Chapter 5: User Experience":
21+
- "Coding Agents":
22+
- "Coding Agents - Nik Pash (Cline)": talks/rag-is-dead-cline-nik.md
23+
- "SWE-Bench Agent - Colin Flaherty (Augment)": talks/colin-rag-agents.md
24+
- "Document Processing":
25+
- "Document Parsing - Adit (Reducto)": talks/reducto-docs-adit.md
26+
- "Browser RAG - Michael (OpenBB)": talks/rag-without-apis-browser-michael-struwig.md
27+
- "Search Technologies":
28+
- "Semantic Search - Will Bryk (Exa)": talks/semantic-search-exa-will-bryk.md
29+
- "RAG Anti-patterns - Skylar Payne": talks/rag-antipatterns-skylar-payne.md
30+
```
31+
32+
## Guidelines
33+
- Keep navigation titles shorter than full article titles for better UX
34+
- Include company names in parentheses when available
35+
- Group related talks into logical subsections
36+
- Maintain consistent formatting across all talk entries
37+
- Use descriptive topic names that indicate the main focus
38+
39+
Reference [AGENTS.md](mdc:docs/talks/AGENTS.md) for complete talk organization guidelines.
Lines changed: 31 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -0,0 +1,31 @@
1+
# Systematically Improving RAG Project Structure
2+
3+
This is a documentation and workshop project for RAG (Retrieval-Augmented Generation) applications.
4+
5+
## Key Directories
6+
- **`docs/`** - Main documentation using MkDocs
7+
- **`docs/talks/`** - Industry expert talks and presentations
8+
- **`docs/workshops/`** - Workshop chapters and exercises
9+
- **`docs/office-hours/`** - Office hours summaries and FAQs
10+
- **`latest/`** - Current cohort materials and case studies
11+
- **`cohort_1/`** & **`cohort_2/`** - Previous cohort materials
12+
- **`data/`** - Sample datasets and examples
13+
14+
## Important Files
15+
- **[mkdocs.yml](mdc:mkdocs.yml)** - Site navigation and configuration
16+
- **[docs/talks/AGENTS.md](mdc:docs/talks/AGENTS.md)** - Talk formatting guidelines
17+
- **[docs/talks/index.md](mdc:docs/talks/index.md)** - Talk directory index
18+
- **[pyproject.toml](mdc:pyproject.toml)** - Python dependencies using `uv`
19+
20+
## User Preferences
21+
- Always use `uv` instead of `pip` for Python package management
22+
- Write at 9th-grade reading level
23+
- Use async over synchronous Python code when possible
24+
- Match existing document styles when writing
25+
- No emojis in documentation
26+
27+
## Build System
28+
- Uses MkDocs for documentation site generation
29+
- Deployed to GitHub Pages
30+
- Run `mkdocs serve` for local development
31+
- Run `mkdocs build` for production builds

.cursor/rules/talks-formatting.mdc

Lines changed: 47 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -0,0 +1,47 @@
1+
# Talks Formatting Standards
2+
3+
When working with files in the `docs/talks/` directory, follow these formatting standards:
4+
5+
## Title Format
6+
All talk titles must follow a **catchy, conversational format** with these principles:
7+
- Use conversational tone with "I", "You", "Why", "How" to make it personal
8+
- Include specific benefits, numbers, or percentages when possible
9+
- Reference the company/organization for credibility
10+
- Use controversial hooks that challenge conventional wisdom
11+
- Suggest actionable implications for readers
12+
13+
### Title Pattern Examples:
14+
- "Why I Stopped Using RAG for Coding Agents (And You Should Too)"
15+
- "The RAG Mistakes That Are Killing Your AI (Lessons from Google & LinkedIn)"
16+
- "Stop Trusting MTEB Rankings - Here's How Chroma Actually Tests Embeddings"
17+
- "The 12% RAG Performance Boost You're Missing (LanceDB's Re-ranking Secrets)"
18+
19+
## YAML Frontmatter Requirements
20+
```yaml
21+
---
22+
title: [Catchy conversational title]
23+
speaker: [Name, Company]
24+
cohort: [Number]
25+
description: [Brief description of key insights]
26+
tags: [relevant, tags, including, company, technology]
27+
date: [YYYY-MM-DD]
28+
---
29+
```
30+
31+
## Content Structure
32+
- **H1 title**: Must match YAML frontmatter title exactly
33+
- **Main sections**: Use `##` for primary sections
34+
- **Subsections**: Use `###` for subsections
35+
- **Key takeaways**: Use `**Key Takeaway:**` format for main insights
36+
- **Quotes**: Use `>` blockquotes for speaker quotes
37+
- **Lists**: Use `-` with **bold** labels for bullet points
38+
- **Company attribution**: Always include company names in titles and content
39+
40+
## Writing Style
41+
- 9th-grade reading level for accessibility
42+
- Technical depth with practical examples
43+
- Actionable insights over theoretical concepts
44+
- Conversational tone that matches catchy titles
45+
- Specific metrics and performance improvements noted
46+
47+
Reference [AGENTS.md](mdc:docs/talks/AGENTS.md) for complete formatting guidelines.

docs/talks/AGENTS.md

Lines changed: 47 additions & 14 deletions
Original file line numberDiff line numberDiff line change
@@ -4,35 +4,68 @@
44
This directory contains industry talks and presentations from the Systematically Improving RAG Applications series. Each talk provides insights from experts at companies like ChromaDB, Zapier, Glean, Exa, and others, covering practical RAG implementation strategies and lessons learned.
55

66
## File Structure
7-
- **Industry expert talks**: 15 markdown files covering specific RAG topics
7+
- **Industry expert talks**: 15+ markdown files covering specific RAG topics
88
- **Organized by chapter**: Talks align with workshop chapters (evaluation, training, UX, etc.)
9-
- **Consistent format**: YAML frontmatter with title, description, tags, speakers, and date
9+
- **Consistent format**: YAML frontmatter with catchy titles, descriptions, tags, speakers, and dates
1010
- **Study notes format**: Key takeaways and technical insights highlighted
1111

12+
## Title Format Standards
13+
All talk titles follow a **catchy, conversational format** designed to grab attention and communicate value:
14+
15+
### Title Pattern Examples:
16+
- **"Why I Stopped Using RAG for Coding Agents (And You Should Too)"** - Personal story + actionable advice
17+
- **"The RAG Mistakes That Are Killing Your AI (Lessons from Google & LinkedIn)"** - Problem identification + company credibility
18+
- **"Stop Trusting MTEB Rankings - Here's How Chroma Actually Tests Embeddings"** - Contrarian take + insider knowledge
19+
- **"The 12% RAG Performance Boost You're Missing (LanceDB's Re-ranking Secrets)"** - Specific benefit + insider secrets
20+
21+
### Title Principles:
22+
- **Conversational tone**: Use "I", "You", "Why", "How" to make it personal
23+
- **Specific benefits**: Include numbers, percentages, or concrete outcomes when possible
24+
- **Company attribution**: Reference the company/organization for credibility
25+
- **Controversial hooks**: Challenge conventional wisdom or common practices
26+
- **Actionable implications**: Suggest there's something readers should do differently
27+
1228
## Content Standards
13-
- **YAML frontmatter**: title, description, tags, speakers, date
14-
- **Study notes structure**: Technical insights with `***Key Takeaway:***` summaries
29+
- **YAML frontmatter**: catchy title, speaker with company, description, tags, date
30+
- **H1 title**: Matches the YAML title exactly for consistency
31+
- **Study notes structure**: Technical insights with `**Key Takeaway:**` summaries
1532
- **Question-answer format**: Practical insights organized by topic
1633
- **Code examples**: Where applicable, include implementation details
1734
- **Performance metrics**: Specific numbers and improvements mentioned
1835

1936
## Key Topics Covered
20-
- **Text chunking strategies** (ChromaDB - Anton)
21-
- **Feedback systems** (Zapier - Vitor)
22-
- **Embedding performance** (Kelly Hong)
23-
- **Enterprise search** (Glean - Manav)
24-
- **Query routing** (Anton)
25-
- **RAG anti-patterns** (Skylar Payne)
26-
- **Semantic search** (Exa - Will Bryk)
27-
- **Online evaluations** (Ben & Sidhant)
28-
- **Browser-based RAG** (Michael Struwig)
37+
- **"Why I Stopped Using RAG for Coding Agents (And You Should Too)"** - Nik Pash (Cline)
38+
- **"The RAG Mistakes That Are Killing Your AI (Lessons from Google & LinkedIn)"** - Skylar Payne
39+
- **"Stop Trusting MTEB Rankings - Here's How Chroma Actually Tests Embeddings"** - Kelly Hong (Chroma)
40+
- **"Why Glean Builds Custom Embedding Models for Every Customer"** - Manav (Glean)
41+
- **"Why Google Search Sucks for AI (And How Exa Is Building Something Better)"** - Will Bryk (Exa)
42+
- **"Why Your AI Is Failing in Production (Lessons from Raindrop & Oleve)"** - Ben & Sidhant
43+
- **"How OpenBB Ditched APIs and Put RAG in the Browser Instead"** - Michael (OpenBB)
44+
- **"Why Grep Beat Embeddings in Our SWE-Bench Agent (Lessons from Augment)"** - Colin Flaherty
45+
- **"Why Most Document Parsing Sucks (And How Reducto Fixed It)"** - Adit (Reducto)
46+
- **"The 12% RAG Performance Boost You're Missing (LanceDB's Re-ranking Secrets)"** - Ayush (LanceDB)
2947

3048
## Writing Style
3149
- **9th-grade reading level** for accessibility
3250
- **Technical depth** with practical examples
3351
- **Actionable insights** over theoretical concepts
3452
- **Surprising discoveries** and counter-intuitive findings highlighted
3553
- **Specific metrics** and performance improvements noted
54+
- **Conversational tone** that matches the catchy titles
55+
56+
## Formatting Standards
57+
- **Consistent H1 titles**: Match YAML frontmatter exactly
58+
- **Proper markdown structure**: Use ## for main sections, ### for subsections
59+
- **Bold key takeaways**: `**Key Takeaway:**` format for main insights
60+
- **Blockquotes for quotes**: Use `>` for speaker quotes
61+
- **Bullet points**: Use `-` for lists with **bold** labels
62+
- **Company attribution**: Always include company names in titles and content
3663

3764
## Tags and Organization
38-
Common tags include: text chunking, evaluation, feedback systems, embedding models, query routing, performance optimization, user experience, production monitoring
65+
Common tags include: RAG, coding agents, embeddings, evaluation, feedback systems, enterprise search, query routing, performance optimization, user experience, production monitoring, document parsing, fine-tuning, re-ranking
66+
67+
---
68+
69+
IF you want to get discounts and 6 day email source on the topic make sure to subscribe to
70+
71+
<script async data-uid="010fd9b52b" src="https://fivesixseven.kit.com/010fd9b52b/index.js"></script>

0 commit comments

Comments
 (0)