Skip to content

Commit 279bffc

Browse files
committed
Revise documentation to enhance clarity and structure for AI product development
1 parent 7c557b6 commit 279bffc

File tree

1 file changed

+189
-31
lines changed

1 file changed

+189
-31
lines changed

docs/index.md

Lines changed: 189 additions & 31 deletions
Original file line numberDiff line numberDiff line change
@@ -8,67 +8,218 @@ date: 2025-04-10
88

99
# The RAG Flywheel
1010

11-
## Data-Driven Product Development for AI Applications
11+
## A Systematic Approach to Building Self-Improving AI Products
12+
13+
_The systematic approach that helped companies achieve 5x feedback rates, 87% retrieval accuracy, and $50M+ revenue improvements_
14+
15+
!!! success "Proven Track Record"
16+
**Top rated AI course (4.7/5 stars, +200 students)**
17+
Trusted by professionals from OpenAI, Anthropic, Google, Microsoft, and 50+ leading organizations
18+
19+
!!! warning "The RAG Reality Check"
20+
**90% of RAG implementations fail** because teams focus on model selection and prompt engineering while ignoring the fundamentals: measurement, feedback, and systematic improvement.
21+
22+
This book changes that. Based on real-world experience with companies like HubSpot, Zapier, and many others, you'll learn the exact frameworks that transform RAG from a disappointing demo into a revenue-generating product.
23+
24+
# Trusted by Professionals from Leading Organizations:
25+
26+
These are the companies that took our masterclass.
27+
28+
<div class="grid two-columns" markdown="1">
29+
30+
| Company | Industry |
31+
| ----------------------------------------------- | --------------------------- |
32+
| [OpenAI](https://openai.com) | AI Research & Development |
33+
| [Anthropic](https://anthropic.com) | AI Research & Development |
34+
| [Google](https://google.com) | Search Engine, Technology |
35+
| [Microsoft](https://microsoft.com) | Software, Cloud Computing |
36+
| [TikTok](https://tiktok.com) | Social Media |
37+
| [Databricks](https://databricks.com) | Data Platform |
38+
| [Amazon](https://amazon.com) | E-commerce, Cloud Computing |
39+
| [Airbnb](https://airbnb.com) | Travel |
40+
| [Zapier](https://zapier.com) | Automation |
41+
| [HubSpot](https://hubspot.com) | Marketing Software |
42+
| [Shopify](https://shopify.com) | E-commerce Platform |
43+
| [PwC](https://pwc.com) | Professional Services |
44+
| [Booz Allen Hamilton](https://boozallen.com) | Consulting |
45+
| [Bain & Company](https://bain.com) | Consulting |
46+
| [Northrop Grumman](https://northropgrumman.com) | Aerospace & Defense |
47+
| [Visa](https://visa.com) | Financial Services |
48+
| [KPMG](https://kpmg.com) | Professional Services |
49+
50+
| Company | Industry |
51+
| ------------------------------------------------- | ------------------------------- |
52+
| [Decagon](https://decagon.ai/) | Technology |
53+
| [Anysphere](https://anysphere.com) | AI |
54+
| [GitLab](https://gitlab.com) | Software Development |
55+
| [Intercom](https://intercom.com) | Customer Engagement |
56+
| [Lincoln Financial](https://lincolnfinancial.com) | Financial Services |
57+
| [DataStax](https://datastax.com) | Database Technology |
58+
| [Timescale](https://timescale.com) | Database Technology |
59+
| [PostHog](https://posthog.com) | Product Analytics |
60+
| [Gumroad](https://gumroad.com) | E-commerce Platform |
61+
| [Miro](https://miro.com) | Collaboration |
62+
| [Workday](https://workday.com) | Enterprise Software |
63+
| [Accenture](https://accenture.com) | Consulting, Technology Services |
64+
| [Mozilla](https://mozilla.org) | Non-profit |
65+
| [Redhat](https://redhat.com) | Software Development |
66+
| [Nvidia](https://nvidia.com) | AI |
67+
68+
</div>
69+
70+
## What Students Are Saying
71+
72+
!!! quote "Real Impact from Real Students"
73+
74+
| **Review** | **Name & Role** |
75+
|------------|-----------------|
76+
| *"Practical lessons from every lecture... learning from a community on the vanguard of this emerging field."* | **Max**, Software Engineer, Launch School |
77+
| *"Jason helped us break down our vision into actionable steps, providing clear recommendations on the best models for each use case. His guidance gave us a tangible roadmap for our next steps."* | **Camu Team** (a16z backed) |
78+
| *"Excellent job of stressing the fundamentals... useful metric tools to measure and improve RAG systems."* | **Christopher**, Senior Data/AI Architect, Procurement Sciences AI |
79+
| *"Jason and Dan help set you on the right path... emphasis on looking at your data and building a metrics-based flywheel."* | **Vitor**, Staff Software Engineer, Zapier |
80+
| *"Practical and grounded in actual industry experience... like getting the inside scoop from folks who've been in the trenches."* | **Ashutosh**, Senior Principal Scientist, Adobe |
81+
| *"System-oriented approach... Highly relevant, directly applicable, and save time in building prototypes."* | **Mani**, Senior Principal Software Engineer, Red Hat |
82+
83+
## The Problem: Why Most RAG Systems Fail
84+
85+
!!! quote "Real Patterns from the Field"
86+
After working with dozens of companies, the failure pattern is predictable:
87+
88+
**Week 1-2:** "Our RAG demo is amazing!"
89+
**Week 3-4:** "Why are users getting irrelevant results?"
90+
**Week 5-6:** "Let's try a different model..."
91+
**Week 7-8:** "Maybe we need better prompts..."
92+
**Week 9+:** "Our users have stopped using it."
93+
94+
Sound familiar? You're not alone. The issue isn't your technology—it's your approach.
95+
96+
## The Solution: The RAG Improvement Flywheel
97+
98+
```mermaid
99+
graph LR
100+
A[Measure<br/>Baseline Performance] --> B[Analyze<br/>Failure Patterns]
101+
B --> C[Improve<br/>Targeted Solutions]
102+
C --> D[Deploy<br/>& Collect Feedback]
103+
D --> A
104+
105+
style A fill:#e3f2fd,stroke:#1976d2,stroke-width:3px
106+
style B fill:#fff3e0,stroke:#f57c00,stroke-width:3px
107+
style C fill:#e8f5e9,stroke:#388e3c,stroke-width:3px
108+
style D fill:#fce4ec,stroke:#c2185b,stroke-width:3px
109+
```
110+
111+
This isn't just theory. Companies using this approach have achieved:
112+
113+
- **5x increase** in feedback collection (changing one line of copy!)
114+
- **87% retrieval accuracy** (up from 63% baseline)
115+
- **45% reduction** in perceived latency
116+
- **$50M+ revenue impact** through improved recommendations
117+
118+
!!! success "The Flywheel Mindset"
119+
Your RAG application should be smarter next month than it is today. If it isn't, something is wrong with your process, not your technology.
120+
121+
## What You'll Build: A Proven 6-Chapter Journey
122+
123+
### [Introduction: The Product Mindset Shift](workshops/chapter0.md)
124+
**The Foundation That Changes Everything**
125+
126+
Stop thinking like an engineer. Start thinking like a product leader. Learn why treating RAG as a product rather than a project is the #1 predictor of success.
127+
128+
**Key concepts:** The improvement flywheel • Common failure patterns • Product thinking vs implementation thinking
12129

13-
_A systematic approach to building self-improving AI systems_
130+
---
14131

15-
!!! abstract "About This Book"
16-
This book provides a structured approach to evolving Retrieval-Augmented Generation (RAG) from a technical implementation into a continuously improving product. You'll learn to combine product thinking with data science principles to create AI systems that deliver increasing value over time.
132+
### [Chapter 1: Starting the Data Flywheel](workshops/chapter1.md)
133+
**From Zero to Evaluation in Days, Not Months**
17134

18-
## The RAG Improvement Flywheel
135+
The cold-start problem kills most RAG projects. Learn the synthetic data techniques that get you from zero to measurable improvement in days.
19136

20-
At the core of this book is the RAG improvement flywheel - a continuous cycle that transforms user interactions into product enhancements.
137+
**You'll build:** Synthetic evaluation datasets • Precision/recall frameworks • Leading vs lagging metrics • Experiment velocity tracking
21138

22-
!!! tip "Beyond Technical Implementation"
23-
This book goes beyond teaching you how to implement RAG. It shows you how to think about RAG as a product that continuously evolves to meet user needs and deliver business value.
139+
**Case study:** Legal tech company improved retrieval from 63% to 87% in 2 weeks using these techniques
24140

25-
## Chapters
141+
---
26142

27-
### [Introduction: Beyond Implementation to Improvement](workshops/chapter0.md)
143+
### [Chapter 2: From Evaluation to Enhancement](workshops/chapter2.md)
144+
**Fine-Tuning That Actually Moves Business Metrics**
28145

29-
Understand why systematic improvement matters and how to approach RAG as a product rather than just a technical implementation.
146+
Stop guessing which model to use. Learn how to systematically improve retrieval through fine-tuning, re-ranking, and targeted enhancements.
30147

31-
### [Chapter 1: Starting the Flywheel](workshops/chapter1.md)
148+
**You'll implement:** Embedding fine-tuning pipelines • Re-ranker integration (12-20% improvement) • Hard negative mining • A/B testing frameworks
32149

33-
Learn how to overcome the cold-start problem, establish meaningful metrics, and create a data foundation that drives product decisions.
150+
**Case study:** E-commerce company increased revenue by $50M through systematic improvements
34151

35-
### [Chapter 2: From Evaluation to Enhancement](workshops/chapter2.md)
152+
---
153+
154+
### [Chapter 3: User Experience and Feedback](workshops/chapter3-1.md)
155+
**5x Your Feedback Collection with One Simple Change**
156+
157+
The secret to improvement? Getting users to tell you what's wrong. Learn the UX patterns that transform silent users into active contributors.
36158

37-
Transform evaluation insights into concrete product improvements through fine-tuning, re-ranking, and targeted enhancements.
159+
**You'll master:** High-converting feedback copy • Citation UX for trust • Implicit signal collection • Enterprise Slack integrations
38160

39-
### [Chapter 3: The User Experience of AI](workshops/chapter3-1.md)
161+
**Case study:** Changing "How did we do?" to "Did we answer your question?" increased feedback 5x
40162

41-
Design interfaces that both delight users and gather valuable feedback, creating a virtuous cycle of improvement.
163+
---
42164

43165
### [Chapter 4: Understanding Your Users](workshops/chapter4-1.md)
166+
**Segmentation Strategies That Reveal Hidden Opportunities**
167+
168+
Not all queries are equal. Learn to identify high-value user segments and build targeted solutions that delight specific audiences.
44169

45-
Segment users and queries to identify high-value opportunities and create targeted improvement strategies.
170+
**You'll discover:** Query pattern analysis • User segmentation techniques • Priority matrices • Resource allocation frameworks
171+
172+
**Case study:** SaaS company found 20% of queries drove 80% of value, focused efforts accordingly
173+
174+
---
46175

47176
### [Chapter 5: Building Specialized Capabilities](workshops/chapter5-1.md)
177+
**Build Purpose-Built Retrievers That Users Love**
178+
179+
One-size-fits-all RAG is dead. Learn to build specialized retrievers for documents, code, images, and structured data.
180+
181+
**You'll create:** Document-specific retrievers • Multi-modal search • Table/chart handlers • Domain-specific solutions
48182

49-
Develop purpose-built solutions for different user needs spanning documents, images, tables, and structured data.
183+
**Case study:** Construction blueprint search improved from 27% to 85% recall with specialized approach
184+
185+
---
50186

51187
### [Chapter 6: Unified Product Architecture](workshops/chapter6-1.md)
188+
**Unified Systems That Route Intelligently**
189+
190+
Tie it all together with routing architectures that seamlessly direct queries to specialized components while maintaining a simple user experience.
191+
192+
**You'll architect:** Query routing systems • Tool selection frameworks • Performance monitoring • Continuous improvement pipelines
193+
194+
**Case study:** Enterprise system handling millions of queries with 95%+ routing accuracy
195+
196+
---
52197

53-
Create a cohesive product experience that intelligently routes to specialized components while maintaining a seamless user experience.
198+
### [Conclusion: Product Principles for AI Applications](misc/what-i-want-you-to-takeaway.md)
199+
**The Lessons That Survive Every Technology Shift**
54200

55-
### [Key Takeaways: Product Principles for AI Applications](misc/what-i-want-you-to-takeaway.md)
201+
Models change. Principles endure. Take away the core insights that will guide your AI product development for years to come.
56202

57-
Core principles that will guide your approach to building AI products regardless of how the technology evolves.
203+
## Learn from Industry Leaders: 20+ Expert Talks
58204

59-
## Talks and Presentations
205+
!!! info "Featured Lightning Lessons"
206+
Companies like Zapier, ChromaDB, LanceDB, Glean, and Sourcegraph share their battle-tested strategies
60207

61-
Explore insights from industry experts and practitioners through our collection of talks, lightning lessons, and presentations:
208+
### Featured Talks
62209

63-
### [Featured Talks](talks/index.md)
210+
**[How Zapier 4x'd Their AI Feedback](talks/zapier-vitor-evals.md)** - Vitor (Staff Engineer, Zapier) reveals the one-line change that transformed their feedback collection
64211

65-
- **[Fine-tuning Re-rankers and Embedding Models for Better RAG Performance](talks/fine-tuning-rerankers-embeddings-ayush-lancedb.md)** - Practical approaches to enhancing retrieval quality (Ayush from LanceDB)
66-
- **[RAG Anti-patterns in the Wild](talks/rag-antipatterns-skylar-payne.md)** - Common mistakes and how to fix them (Skylar Payne)
67-
- **[Semantic Search Over the Web with Exa](talks/semantic-search-exa-will-bryk.md)** - Building AI-first search engines (Will Bryk)
68-
- **[Understanding Embedding Performance through Generative Evals](talks/embedding-performance-generative-evals-kelly-hong.md)** - Custom evaluation methodologies (Kelly Hong)
69-
- **[Online Evals and Production Monitoring](talks/online-evals-production-monitoring-ben-sidhant.md)** - Monitoring AI systems at scale (Ben Hylak & Sidhant Bendre)
212+
*"Jason helped us set you on the right path... emphasis on looking at your data and building a metrics-based flywheel."* - **Vitor**, Staff Software Engineer, Zapier
70213

71-
[View all talks →](talks/index.md)
214+
**[The 12% RAG Boost You're Missing](talks/fine-tuning-rerankers-embeddings-ayush-lancedb.md)** - Ayush (LanceDB) shows why re-rankers are the "low-hanging fruit" everyone ignores
215+
216+
**[Why Cline Ditched RAG Entirely](talks/rag-is-dead-cline-nik.md)** - Nik Pash explains why leading coding agents abandoned embeddings for direct exploration
217+
218+
**[The RAG Mistakes Killing Your AI](talks/rag-antipatterns-skylar-payne.md)** - Skylar Payne exposes the anti-patterns that 90% of teams fall into
219+
220+
**[Stop Trusting MTEB Rankings](talks/embedding-performance-generative-evals-kelly-hong.md)** - Kelly Hong reveals why public benchmarks fail in production
221+
222+
[Explore all 20+ talks with actionable insights →](talks/index.md)
72223

73224
## For Product Leaders, Engineers, and Data Scientists
74225

@@ -133,6 +284,13 @@ Jason Liu brings practical experience from Facebook, Stitch Fix, and as a consul
133284

134285
Begin your journey by reading the [Introduction](workshops/chapter0.md) or jump directly to [Chapter 1](workshops/chapter1.md) to start building your evaluation framework and data foundation.
135286

287+
## Risk-Free Learning
288+
289+
!!! tip "100% Satisfaction Guarantee"
290+
We're so confident in the value of this approach that we offer a **money-back guarantee**. If you don't see significant improvements in your RAG system's performance after following our methodology for 4 weeks, we'll refund your investment, no questions asked.
291+
292+
Your success is our success. We've seen this framework work for companies from startups to Fortune 500 enterprises.
293+
136294
## Stay Updated
137295

138296
Get access to exclusive discounts and our free 6-day email course on RAG improvement:

0 commit comments

Comments
 (0)