From 8712b1352880f7855e1d0f4a51675d47dec9a4a2 Mon Sep 17 00:00:00 2001 From: Jason Liu Date: Wed, 1 Oct 2025 11:42:54 -0400 Subject: [PATCH] chore: remove enrollment button snippet --- README.md | 4 ---- cohort_1/README.md | 5 ----- cohort_1/week1_bootstrap_evals/README.md | 5 ----- .../classify_with_yaml/readme.md | 5 ----- .../example_notebook/README.md | 5 ----- cohort_1/week3_seperate_indices/README.md | 5 ----- cohort_1/week4_routing/README.md | 5 ----- cohort_1/week5_fine_tuning/README.md | 5 ----- cohort_2/README.md | 5 ----- cohort_2/office-hours/README.md | 5 ----- cohort_2/office-hours/week1-summary.md | 4 ---- cohort_2/office-hours/week2-summary.md | 5 ----- cohort_2/office-hours/week3-summary.md | 4 ---- cohort_2/office-hours/week3/week3-summary.md | 4 ---- cohort_2/office-hours/week4-summary.md | 5 ----- cohort_2/week0/README.md | 5 ----- cohort_2/week0/uv.md | 5 ----- cohort_2/week1/README.md | 5 ----- cohort_2/week2/README.md | 4 ---- cohort_2/week4/README.md | 4 ---- cohort_2/week5/README.md | 4 ---- cohort_2/week6/README.md | 4 ---- docs/blog.md | 5 ----- docs/office-hours/AGENTS.md | 5 ----- docs/office-hours/cohort2/week1-summary.md | 5 ----- docs/office-hours/cohort2/week2-summary.md | 5 ----- docs/office-hours/cohort2/week3-summary.md | 5 ----- docs/office-hours/cohort2/week4-summary.md | 5 ----- docs/office-hours/cohort2/week5-summary.md | 5 ----- docs/office-hours/cohort2/week6-summary.md | 5 ----- docs/office-hours/cohort3/week-4-2.md | 5 ----- docs/office-hours/cohort3/week-5-1.md | 5 ----- docs/office-hours/cohort3/week-5-2.md | 5 ----- docs/office-hours/faq.md | 6 ------ docs/snippets/README.md | 3 --- docs/talks/AGENTS.md | 5 ----- docs/talks/chromadb-anton-chunking.md | 4 ---- docs/talks/colin-rag-agents.md | 4 ---- ...performance-generative-evals-kelly-hong.md | 5 ----- docs/talks/extend-document-automation.md | 5 ----- ...ning-rerankers-embeddings-ayush-lancedb.md | 4 ---- docs/talks/glean-manav.md | 6 ------ docs/talks/john-lexical-search.md | 4 ---- ...evals-production-monitoring-ben-sidhant.md | 5 ----- docs/talks/query-routing-anton.md | 4 ---- docs/talks/rag-antipatterns-skylar-payne.md | 5 ----- docs/talks/rag-is-dead-cline-nik.md | 5 ----- ...ag-without-apis-browser-michael-struwig.md | 5 ----- docs/talks/reducto-docs-adit.md | 5 ----- docs/talks/semantic-search-exa-will-bryk.md | 3 --- docs/talks/superlinked-encoder-stacking.md | 4 ---- docs/talks/zapier-vitor-evals.md | 4 ---- latest/README.md | 5 ----- latest/README_TEMPLATE.md | 4 ---- latest/case_study/CLAUDE.md | 4 ---- latest/case_study/README.md | 5 ----- latest/case_study/teaching/part01/README.md | 4 ---- latest/case_study/teaching/part02/README.md | Bin 16763 -> 16712 bytes .../teaching/part02/analysis_summary_1000.md | 4 ---- latest/case_study/teaching/part03/README.md | 4 ---- latest/case_study/teaching/part04/README.md | 5 ----- latest/week0/README.md | 4 ---- latest/week1/README.md | 5 ----- latest/week2/README.md | 4 ---- latest/week4/README.md | 4 ---- latest/week5/README.md | 4 ---- latest/week6/README.md | 4 ---- md/extra_kura_01_cluster_conversations.md | 5 ----- md/extra_kura_02_better_summaries.md | 5 ----- md/extra_kura_03_classifiers.md | 5 ----- md/week6_03_improving_performance.md | 5 ----- md/week6_03_improving_performance_logfire.md | 5 ----- 72 files changed, 328 deletions(-) diff --git a/README.md b/README.md index c3434a3..4c40c93 100644 --- a/README.md +++ b/README.md @@ -251,8 +251,4 @@ Want to learn more about RAG? Take our free email course and get the latest news **Join thousands of developers who are building better AI applications with our proven methodology.** ---8<-- -"snippets/enrollment-button.md" ---8<-- - _Learn evaluation strategies, user experience design, and architectural patterns that actually work in production._ diff --git a/cohort_1/README.md b/cohort_1/README.md index 80f2896..bfd47d0 100644 --- a/cohort_1/README.md +++ b/cohort_1/README.md @@ -30,8 +30,3 @@ The `latest/` directory contains: --- ---8<-- -"snippets/enrollment-button.md" ---8<-- - ---- diff --git a/cohort_1/week1_bootstrap_evals/README.md b/cohort_1/week1_bootstrap_evals/README.md index 80f02dc..1123053 100644 --- a/cohort_1/week1_bootstrap_evals/README.md +++ b/cohort_1/week1_bootstrap_evals/README.md @@ -10,8 +10,3 @@ Finally, we calculate metrics on these questions in `metrics.ipynb`. This uses a --- ---8<-- -"snippets/enrollment-button.md" ---8<-- - ---- diff --git a/cohort_1/week2_question_classification/classify_with_yaml/readme.md b/cohort_1/week2_question_classification/classify_with_yaml/readme.md index 9260a0f..9a6aebf 100644 --- a/cohort_1/week2_question_classification/classify_with_yaml/readme.md +++ b/cohort_1/week2_question_classification/classify_with_yaml/readme.md @@ -30,8 +30,3 @@ The `RAGClassifier` is a classifier that uses a retreival model to classify user --- ---8<-- -"snippets/enrollment-button.md" ---8<-- - ---- diff --git a/cohort_1/week2_question_classification/example_notebook/README.md b/cohort_1/week2_question_classification/example_notebook/README.md index 22dee73..142559b 100644 --- a/cohort_1/week2_question_classification/example_notebook/README.md +++ b/cohort_1/week2_question_classification/example_notebook/README.md @@ -8,8 +8,3 @@ Because you'll want data when first exploring this code, the notebook uses synth --- ---8<-- -"snippets/enrollment-button.md" ---8<-- - ---- diff --git a/cohort_1/week3_seperate_indices/README.md b/cohort_1/week3_seperate_indices/README.md index f31ee0c..46c321a 100644 --- a/cohort_1/week3_seperate_indices/README.md +++ b/cohort_1/week3_seperate_indices/README.md @@ -4,8 +4,3 @@ It continues the example of a hardware retailer building a system to answer user --- ---8<-- -"snippets/enrollment-button.md" ---8<-- - ---- diff --git a/cohort_1/week4_routing/README.md b/cohort_1/week4_routing/README.md index 022f2b0..9cdfc9b 100644 --- a/cohort_1/week4_routing/README.md +++ b/cohort_1/week4_routing/README.md @@ -6,8 +6,3 @@ This directory shows how to test whether we are retrieving the correct tools. It --- ---8<-- -"snippets/enrollment-button.md" ---8<-- - ---- diff --git a/cohort_1/week5_fine_tuning/README.md b/cohort_1/week5_fine_tuning/README.md index 37d1f74..e945469 100644 --- a/cohort_1/week5_fine_tuning/README.md +++ b/cohort_1/week5_fine_tuning/README.md @@ -10,8 +10,3 @@ The key files in this directory are: --- ---8<-- -"snippets/enrollment-button.md" ---8<-- - ---- diff --git a/cohort_2/README.md b/cohort_2/README.md index 6cc1902..92c3ae9 100644 --- a/cohort_2/README.md +++ b/cohort_2/README.md @@ -200,8 +200,3 @@ Want to learn more about RAG? Take our free email course and get the latest news --- ---8<-- -"snippets/enrollment-button.md" ---8<-- - ---- diff --git a/cohort_2/office-hours/README.md b/cohort_2/office-hours/README.md index 8b5f6cf..147c8dd 100644 --- a/cohort_2/office-hours/README.md +++ b/cohort_2/office-hours/README.md @@ -129,8 +129,3 @@ The script handles various transcript file formats and naming patterns, includin --- ---8<-- -"snippets/enrollment-button.md" ---8<-- - ---- diff --git a/cohort_2/office-hours/week1-summary.md b/cohort_2/office-hours/week1-summary.md index e2c139c..42424b8 100644 --- a/cohort_2/office-hours/week1-summary.md +++ b/cohort_2/office-hours/week1-summary.md @@ -324,8 +324,4 @@ Even with one-to-one mappings, MRR (Mean Reciprocal Rank) is still useful to see ## What really matters isn't the absolute number but whether we can move these metrics in a positive direction with our interventions. It's like weighing yourself—the absolute number may vary by scale, but if you've gained two pounds, you've definitely gained two pounds. ---8<-- -"snippets/enrollment-button.md" ---8<-- - --- diff --git a/cohort_2/office-hours/week2-summary.md b/cohort_2/office-hours/week2-summary.md index 7577ba7..1329902 100644 --- a/cohort_2/office-hours/week2-summary.md +++ b/cohort_2/office-hours/week2-summary.md @@ -332,8 +332,3 @@ This approach helps ensure reliability across different types of function callin --- ---8<-- -"snippets/enrollment-button.md" ---8<-- - ---- diff --git a/cohort_2/office-hours/week3-summary.md b/cohort_2/office-hours/week3-summary.md index 8dfaa02..a9d4696 100644 --- a/cohort_2/office-hours/week3-summary.md +++ b/cohort_2/office-hours/week3-summary.md @@ -258,8 +258,4 @@ Even systems like Deep Research might evolve to pull from sources you tend to ag ## _Note: I'll continue to add resources and notes from future office hours sessions_ ---8<-- -"snippets/enrollment-button.md" ---8<-- - --- diff --git a/cohort_2/office-hours/week3/week3-summary.md b/cohort_2/office-hours/week3/week3-summary.md index 8dfaa02..a9d4696 100644 --- a/cohort_2/office-hours/week3/week3-summary.md +++ b/cohort_2/office-hours/week3/week3-summary.md @@ -258,8 +258,4 @@ Even systems like Deep Research might evolve to pull from sources you tend to ag ## _Note: I'll continue to add resources and notes from future office hours sessions_ ---8<-- -"snippets/enrollment-button.md" ---8<-- - --- diff --git a/cohort_2/office-hours/week4-summary.md b/cohort_2/office-hours/week4-summary.md index e1049af..7003a0d 100644 --- a/cohort_2/office-hours/week4-summary.md +++ b/cohort_2/office-hours/week4-summary.md @@ -347,8 +347,3 @@ The protocol is still evolving but shows promise for creating more powerful AI s --- ---8<-- -"snippets/enrollment-button.md" ---8<-- - ---- diff --git a/cohort_2/week0/README.md b/cohort_2/week0/README.md index 6680e2f..1cf4c42 100644 --- a/cohort_2/week0/README.md +++ b/cohort_2/week0/README.md @@ -62,8 +62,3 @@ These notebooks provide the essential tools and concepts you'll use throughout t --- ---8<-- -"snippets/enrollment-button.md" ---8<-- - ---- diff --git a/cohort_2/week0/uv.md b/cohort_2/week0/uv.md index 65f2ec8..2ecfcbd 100644 --- a/cohort_2/week0/uv.md +++ b/cohort_2/week0/uv.md @@ -104,8 +104,3 @@ When working with virtual environments, we recommend the following best practice --- ---8<-- -"snippets/enrollment-button.md" ---8<-- - ---- diff --git a/cohort_2/week1/README.md b/cohort_2/week1/README.md index 7d2917e..e7e8ca1 100644 --- a/cohort_2/week1/README.md +++ b/cohort_2/week1/README.md @@ -82,8 +82,3 @@ This approach ensures you invest in retrieval techniques that provide measurable --- ---8<-- -"snippets/enrollment-button.md" ---8<-- - ---- diff --git a/cohort_2/week2/README.md b/cohort_2/week2/README.md index 3cc1377..debf7f9 100644 --- a/cohort_2/week2/README.md +++ b/cohort_2/week2/README.md @@ -71,8 +71,4 @@ These notebooks demonstrate two distinct approaches to improving embedding perfo ## Both approaches show significant performance gains (15-30% in MRR and recall), highlighting that domain-specific fine-tuning is one of the most effective ways to improve RAG system performance. ---8<-- -"snippets/enrollment-button.md" ---8<-- - --- diff --git a/cohort_2/week4/README.md b/cohort_2/week4/README.md index 742d5a4..6958272 100644 --- a/cohort_2/week4/README.md +++ b/cohort_2/week4/README.md @@ -72,8 +72,4 @@ Together, these notebooks provide a systematic approach to understanding and imp ## This data-driven approach helps identify which query areas need improvement most urgently, allowing targeted enhancements rather than making random changes. ---8<-- -"snippets/enrollment-button.md" ---8<-- - --- diff --git a/cohort_2/week5/README.md b/cohort_2/week5/README.md index 80ea8b4..9323819 100644 --- a/cohort_2/week5/README.md +++ b/cohort_2/week5/README.md @@ -90,8 +90,4 @@ These notebooks move beyond simple semantic search to address complex real-world ## By combining vector search with structured filtering, SQL access, and document parsing, you'll build RAG systems capable of handling sophisticated user needs that pure semantic search can't address alone. ---8<-- -"snippets/enrollment-button.md" ---8<-- - --- diff --git a/cohort_2/week6/README.md b/cohort_2/week6/README.md index bdc91a2..f94aaf4 100644 --- a/cohort_2/week6/README.md +++ b/cohort_2/week6/README.md @@ -73,8 +73,4 @@ Modern RAG systems increasingly rely on tool orchestration beyond simple retriev ## This approach demonstrates how simple prompting changes can dramatically improve tool selection, enabling RAG systems to coordinate multiple specialized capabilities rather than relying on retrieval alone. ---8<-- -"snippets/enrollment-button.md" ---8<-- - --- diff --git a/docs/blog.md b/docs/blog.md index bc7c34d..eee11a4 100644 --- a/docs/blog.md +++ b/docs/blog.md @@ -256,8 +256,3 @@ For a deeper dive into these concepts, check out the free 6-week email course on --- ---8<-- -"snippets/enrollment-button.md" ---8<-- - ---- diff --git a/docs/office-hours/AGENTS.md b/docs/office-hours/AGENTS.md index 8021472..fb702f8 100644 --- a/docs/office-hours/AGENTS.md +++ b/docs/office-hours/AGENTS.md @@ -60,8 +60,3 @@ topics: --- ---8<-- -"snippets/enrollment-button.md" ---8<-- - ---- diff --git a/docs/office-hours/cohort2/week1-summary.md b/docs/office-hours/cohort2/week1-summary.md index 8916687..f2ab701 100644 --- a/docs/office-hours/cohort2/week1-summary.md +++ b/docs/office-hours/cohort2/week1-summary.md @@ -354,8 +354,3 @@ What really matters isn't the absolute number but whether we can move these metr --- ---8<-- -"snippets/enrollment-button.md" ---8<-- - ---- diff --git a/docs/office-hours/cohort2/week2-summary.md b/docs/office-hours/cohort2/week2-summary.md index 7e6e902..9d3bf31 100644 --- a/docs/office-hours/cohort2/week2-summary.md +++ b/docs/office-hours/cohort2/week2-summary.md @@ -360,8 +360,3 @@ This approach helps ensure reliability across different types of function callin --- ---8<-- -"snippets/enrollment-button.md" ---8<-- - ---- diff --git a/docs/office-hours/cohort2/week3-summary.md b/docs/office-hours/cohort2/week3-summary.md index 6388dcd..2d5baa3 100644 --- a/docs/office-hours/cohort2/week3-summary.md +++ b/docs/office-hours/cohort2/week3-summary.md @@ -288,8 +288,3 @@ _Note: I'll continue to add resources and notes from future office hours session --- ---8<-- -"snippets/enrollment-button.md" ---8<-- - ---- diff --git a/docs/office-hours/cohort2/week4-summary.md b/docs/office-hours/cohort2/week4-summary.md index 48788ce..9584c4f 100644 --- a/docs/office-hours/cohort2/week4-summary.md +++ b/docs/office-hours/cohort2/week4-summary.md @@ -372,8 +372,3 @@ The protocol is still evolving but shows promise for creating more powerful AI s --- ---8<-- -"snippets/enrollment-button.md" ---8<-- - ---- diff --git a/docs/office-hours/cohort2/week5-summary.md b/docs/office-hours/cohort2/week5-summary.md index 6248444..5db473b 100644 --- a/docs/office-hours/cohort2/week5-summary.md +++ b/docs/office-hours/cohort2/week5-summary.md @@ -186,8 +186,3 @@ Follow-up prompts at the end of interactions help maintain momentum. By asking w --- ---8<-- -"snippets/enrollment-button.md" ---8<-- - ---- diff --git a/docs/office-hours/cohort2/week6-summary.md b/docs/office-hours/cohort2/week6-summary.md index c4a4bff..06b564f 100644 --- a/docs/office-hours/cohort2/week6-summary.md +++ b/docs/office-hours/cohort2/week6-summary.md @@ -239,8 +239,3 @@ For most organizations building AI applications, starting with a specific domain --- ---8<-- -"snippets/enrollment-button.md" ---8<-- - ---- diff --git a/docs/office-hours/cohort3/week-4-2.md b/docs/office-hours/cohort3/week-4-2.md index d2e1262..02498b3 100644 --- a/docs/office-hours/cohort3/week-4-2.md +++ b/docs/office-hours/cohort3/week-4-2.md @@ -129,8 +129,3 @@ Beyond the course, I'm organizing a speaker series with guests from OpenAI's mem --- ---8<-- -"snippets/enrollment-button.md" ---8<-- - ---- diff --git a/docs/office-hours/cohort3/week-5-1.md b/docs/office-hours/cohort3/week-5-1.md index f837330..95edef6 100644 --- a/docs/office-hours/cohort3/week-5-1.md +++ b/docs/office-hours/cohort3/week-5-1.md @@ -270,8 +270,3 @@ Newer models with improved attention mechanisms handle long contexts better than --- ---8<-- -"snippets/enrollment-button.md" ---8<-- - ---- diff --git a/docs/office-hours/cohort3/week-5-2.md b/docs/office-hours/cohort3/week-5-2.md index 71eabcf..cfe8ac7 100644 --- a/docs/office-hours/cohort3/week-5-2.md +++ b/docs/office-hours/cohort3/week-5-2.md @@ -197,8 +197,3 @@ Not entirely. Even as models improve at reasoning, having a portfolio of special --- ---8<-- -"snippets/enrollment-button.md" ---8<-- - ---- diff --git a/docs/office-hours/faq.md b/docs/office-hours/faq.md index 846e599..2a33bea 100644 --- a/docs/office-hours/faq.md +++ b/docs/office-hours/faq.md @@ -409,11 +409,6 @@ What really matters isn't the absolute number but whether we can move these metr --- ---8<-- -"snippets/enrollment-button.md" ---8<-- - ---- ## How would you evaluate the effect of different parsing strategies in RAG, notably on documents with weird layouts, tables, and charts? @@ -3921,4 +3916,3 @@ Found an error or want to suggest improvements to these FAQs? The source files a --- ---- diff --git a/docs/snippets/README.md b/docs/snippets/README.md index 01ac527..c4e2bdb 100644 --- a/docs/snippets/README.md +++ b/docs/snippets/README.md @@ -9,9 +9,6 @@ This directory contains reusable content snippets that can be included in any ma Simple enrollment button without additional styling: ```markdown ---8<-- -"snippets/enrollment-button.md" ---8<-- ``` ### 2. `enrollment-section.md` diff --git a/docs/talks/AGENTS.md b/docs/talks/AGENTS.md index 54440a3..25f6a9a 100644 --- a/docs/talks/AGENTS.md +++ b/docs/talks/AGENTS.md @@ -116,8 +116,3 @@ Common tags include: RAG, coding agents, embeddings, evaluation, feedback system --- ---8<-- -"snippets/enrollment-button.md" ---8<-- - ---- diff --git a/docs/talks/chromadb-anton-chunking.md b/docs/talks/chromadb-anton-chunking.md index 26505ce..17489be 100644 --- a/docs/talks/chromadb-anton-chunking.md +++ b/docs/talks/chromadb-anton-chunking.md @@ -107,8 +107,4 @@ The ChromaDB team is developing better tooling to help with this process, but in ## By focusing on these fundamentals rather than blindly applying frameworks or following defaults, you can significantly improve the performance of your RAG applications and deliver better results to your users. ---8<-- -"snippets/enrollment-button.md" ---8<-- - --- diff --git a/docs/talks/colin-rag-agents.md b/docs/talks/colin-rag-agents.md index 26ec4a2..a582f4f 100644 --- a/docs/talks/colin-rag-agents.md +++ b/docs/talks/colin-rag-agents.md @@ -185,8 +185,4 @@ For the SWE-Bench coding evaluation, embedding models didn't significantly impro ## This doesn't mean embedding models aren't valuable—they become essential for larger codebases, less structured content, or more complex retrieval tasks. ---8<-- -"snippets/enrollment-button.md" ---8<-- - --- diff --git a/docs/talks/embedding-performance-generative-evals-kelly-hong.md b/docs/talks/embedding-performance-generative-evals-kelly-hong.md index 725c226..ffc82f1 100644 --- a/docs/talks/embedding-performance-generative-evals-kelly-hong.md +++ b/docs/talks/embedding-performance-generative-evals-kelly-hong.md @@ -190,8 +190,3 @@ Contextual chunk rewriting involves adding context to document chunks to improve --- ---8<-- -"snippets/enrollment-button.md" ---8<-- - ---- diff --git a/docs/talks/extend-document-automation.md b/docs/talks/extend-document-automation.md index 1e1bd06..690e033 100644 --- a/docs/talks/extend-document-automation.md +++ b/docs/talks/extend-document-automation.md @@ -249,8 +249,3 @@ This approach delivers business value faster while building organizational comfo --- ---8<-- -"snippets/enrollment-button.md" ---8<-- - ---- diff --git a/docs/talks/fine-tuning-rerankers-embeddings-ayush-lancedb.md b/docs/talks/fine-tuning-rerankers-embeddings-ayush-lancedb.md index 6ba63d2..8caea60 100644 --- a/docs/talks/fine-tuning-rerankers-embeddings-ayush-lancedb.md +++ b/docs/talks/fine-tuning-rerankers-embeddings-ayush-lancedb.md @@ -231,8 +231,4 @@ Start with a small experiment using a lightweight model that you can train quick ## Multimodal retrieval is likely the next frontier. While much work has been done on text retrieval, there's still significant room for improvement in retrieving and ranking content across different modalities like images, audio, and video. Building better benchmarks and baselines for multimodal RAG systems represents an important area for future development. ---8<-- -"snippets/enrollment-button.md" ---8<-- - --- diff --git a/docs/talks/glean-manav.md b/docs/talks/glean-manav.md index ee2da78..cd5f8e6 100644 --- a/docs/talks/glean-manav.md +++ b/docs/talks/glean-manav.md @@ -182,10 +182,4 @@ Glean's approach is refreshingly pragmatic. They've learned that the path to hig The emphasis on company-specific language models is particularly noteworthy - this is an area where many companies struggle when they try to apply generic embedding models to their unique terminology and document structures. **Key Takeaway:** Successful enterprise search requires a pragmatic approach that combines custom embedding models, unified data architecture, hybrid search techniques, and continuous learning from user feedback rather than relying solely on off-the-shelf solutions. - -## --8<-- - -"snippets/enrollment-button.md" ---8<-- - --- diff --git a/docs/talks/john-lexical-search.md b/docs/talks/john-lexical-search.md index 41421cd..a392762 100644 --- a/docs/talks/john-lexical-search.md +++ b/docs/talks/john-lexical-search.md @@ -115,8 +115,4 @@ The most promising direction appears to be hybrid approaches that leverage the s ## **_Key Takeaway:_** The debate isn't really about lexical versus semantic search, but rather how to effectively combine them to create retrieval systems that are both precise and understanding - offering both the filtering power of traditional search and the semantic comprehension of modern embedding-based approaches. ---8<-- -"snippets/enrollment-button.md" ---8<-- - --- diff --git a/docs/talks/online-evals-production-monitoring-ben-sidhant.md b/docs/talks/online-evals-production-monitoring-ben-sidhant.md index 8cb6ad3..0d7309a 100644 --- a/docs/talks/online-evals-production-monitoring-ben-sidhant.md +++ b/docs/talks/online-evals-production-monitoring-ben-sidhant.md @@ -224,8 +224,3 @@ Monitor your system before and after making changes to see if the frequency of r --- ---8<-- -"snippets/enrollment-button.md" ---8<-- - ---- diff --git a/docs/talks/query-routing-anton.md b/docs/talks/query-routing-anton.md index f032074..ebb1293 100644 --- a/docs/talks/query-routing-anton.md +++ b/docs/talks/query-routing-anton.md @@ -166,8 +166,4 @@ The key message throughout Anton's presentation was that data organization isn't ## **_Key Takeaway:_** Data organization for RAG systems requires thinking beyond traditional search paradigms. By carefully considering user access patterns, data source characteristics, and query requirements, you can design a system that delivers better results while maintaining security and performance. ---8<-- -"snippets/enrollment-button.md" ---8<-- - --- diff --git a/docs/talks/rag-antipatterns-skylar-payne.md b/docs/talks/rag-antipatterns-skylar-payne.md index 24541fd..291cf65 100644 --- a/docs/talks/rag-antipatterns-skylar-payne.md +++ b/docs/talks/rag-antipatterns-skylar-payne.md @@ -282,8 +282,3 @@ Metadata becomes more valuable when you have both high data volume and diverse q --- ---8<-- -"snippets/enrollment-button.md" ---8<-- - ---- diff --git a/docs/talks/rag-is-dead-cline-nik.md b/docs/talks/rag-is-dead-cline-nik.md index c93197f..0423fc8 100644 --- a/docs/talks/rag-is-dead-cline-nik.md +++ b/docs/talks/rag-is-dead-cline-nik.md @@ -227,8 +227,3 @@ There's a balance between providing specialized tools (read file, search directo --- ---8<-- -"snippets/enrollment-button.md" ---8<-- - ---- diff --git a/docs/talks/rag-without-apis-browser-michael-struwig.md b/docs/talks/rag-without-apis-browser-michael-struwig.md index 003bf0d..77a139f 100644 --- a/docs/talks/rag-without-apis-browser-michael-struwig.md +++ b/docs/talks/rag-without-apis-browser-michael-struwig.md @@ -267,8 +267,3 @@ OpenBB believes that human plus AI will be better than either human or AI alone. --- ---8<-- -"snippets/enrollment-button.md" ---8<-- - ---- diff --git a/docs/talks/reducto-docs-adit.md b/docs/talks/reducto-docs-adit.md index bab4b70..53e490a 100644 --- a/docs/talks/reducto-docs-adit.md +++ b/docs/talks/reducto-docs-adit.md @@ -322,8 +322,3 @@ Agentic RAG involves adding reasoning capabilities to determine what information --- ---8<-- -"snippets/enrollment-button.md" ---8<-- - ---- diff --git a/docs/talks/semantic-search-exa-will-bryk.md b/docs/talks/semantic-search-exa-will-bryk.md index 4f0adc5..d8b4b8c 100644 --- a/docs/talks/semantic-search-exa-will-bryk.md +++ b/docs/talks/semantic-search-exa-will-bryk.md @@ -203,6 +203,3 @@ Exa's neural embedding-based approach makes it inherently more resistant to trad Exa works with web documents which naturally have a structure that makes them manageable for embedding. While there are limits to document size, Exa can handle most web content effectively. The company is exploring chunking strategies for longer documents but has found that the natural structure of web content often provides reasonable document boundaries. ---8<-- -"snippets/enrollment-button.md" ---8<-- diff --git a/docs/talks/superlinked-encoder-stacking.md b/docs/talks/superlinked-encoder-stacking.md index d186cdf..11d8e69 100644 --- a/docs/talks/superlinked-encoder-stacking.md +++ b/docs/talks/superlinked-encoder-stacking.md @@ -175,8 +175,4 @@ While large companies with massive data might be able to train transformers that ## We should focus on three key areas: 1) Looking at the world through lenses beyond text tokenization, 2) Reducing our reliance on re-ranking by improving initial retrieval, and 3) Using smooth bias functions instead of rigid Boolean filters. These approaches will lead to more accurate, efficient, and controllable search systems. ---8<-- -"snippets/enrollment-button.md" ---8<-- - --- diff --git a/docs/talks/zapier-vitor-evals.md b/docs/talks/zapier-vitor-evals.md index 55e28d7..b587603 100644 --- a/docs/talks/zapier-vitor-evals.md +++ b/docs/talks/zapier-vitor-evals.md @@ -116,8 +116,4 @@ This cycle becomes particularly valuable when evaluating new model versions. Whe ## **_Key Takeaway:_** Feedback collection isn't just about fixing immediate issues - it's a critical component of a sustainable AI product development cycle. Each round of feedback strengthens your evaluation suite, which in turn enables more confident decision-making about system changes and model upgrades. ---8<-- -"snippets/enrollment-button.md" ---8<-- - --- diff --git a/latest/README.md b/latest/README.md index d8ff153..2035ca1 100644 --- a/latest/README.md +++ b/latest/README.md @@ -207,8 +207,3 @@ Want to learn more about RAG? Take our free email course and get the latest news --- ---8<-- -"snippets/enrollment-button.md" ---8<-- - ---- diff --git a/latest/README_TEMPLATE.md b/latest/README_TEMPLATE.md index 87eeffe..52a6656 100644 --- a/latest/README_TEMPLATE.md +++ b/latest/README_TEMPLATE.md @@ -107,8 +107,4 @@ After completing this week's materials, you'll have: ## **Note**: Ensure you've completed all prerequisites before starting these notebooks. Each notebook builds on previous concepts. ---8<-- -"snippets/enrollment-button.md" ---8<-- - --- diff --git a/latest/case_study/CLAUDE.md b/latest/case_study/CLAUDE.md index 0da67c5..86db5f1 100644 --- a/latest/case_study/CLAUDE.md +++ b/latest/case_study/CLAUDE.md @@ -322,8 +322,4 @@ This case study demonstrates key concepts from the workshop series: ## Remember: The goal isn't perfect recall, but understanding and solving alignment problems systematically. ---8<-- -"snippets/enrollment-button.md" ---8<-- - --- diff --git a/latest/case_study/README.md b/latest/case_study/README.md index f32af64..714be57 100644 --- a/latest/case_study/README.md +++ b/latest/case_study/README.md @@ -324,8 +324,3 @@ The key insight you'll discover: **In RAG systems, alignment between queries and --- ---8<-- -"snippets/enrollment-button.md" ---8<-- - ---- diff --git a/latest/case_study/teaching/part01/README.md b/latest/case_study/teaching/part01/README.md index 2205434..c33dbac 100644 --- a/latest/case_study/teaching/part01/README.md +++ b/latest/case_study/teaching/part01/README.md @@ -139,8 +139,4 @@ With this understanding of the data distribution, we can: ## _This analysis is based on 9,930 conversations loaded from the WildChat-1M dataset with full text preservation._ ---8<-- -"snippets/enrollment-button.md" ---8<-- - --- diff --git a/latest/case_study/teaching/part02/README.md b/latest/case_study/teaching/part02/README.md index caff481de65b0871413b74359b40e40ff89dc71a..6871e02fd0b49db427fbf0ecd37b3c010fddd761 100644 GIT binary patch delta 91 zcmey}#CW2Kal=k7el7+Eg_OjSM13wM1qB7J$tSqt8M!7m@<>m1<~9bh%DDsBKx)bi nCkOHfPkz9i3>0+XnZOJboZQbZHu)`&JeXZzAii0RSIGnbW33ks delta 143 zcmX@n#Q3|3al=k7ei;S^g_OjSM13wM1qB6}$tSqt8D%Cn@<>m1<~9bh%DDsBKx)bi zCkOHfPkz9i3>0+XnZOJboZQbZHu)`&JeXZzAii0RSINX!SJ%QuSC>nvI4`rHAho1e hKQ*r?KPM+QHLpZBskEddKTj_=MTrZl3`py80RZ^DCd&W- diff --git a/latest/case_study/teaching/part02/analysis_summary_1000.md b/latest/case_study/teaching/part02/analysis_summary_1000.md index fe2a6a1..c049d30 100644 --- a/latest/case_study/teaching/part02/analysis_summary_1000.md +++ b/latest/case_study/teaching/part02/analysis_summary_1000.md @@ -69,8 +69,4 @@ We evaluated 3 embedding models on 1000 conversations from the WildChat dataset, ## This experiment proves that **alignment beats model sophistication**. A 384-dimensional model with good alignment (V1) outperforms a 3072-dimensional model with poor alignment (V2) by over 40%. Focus on alignment before investing in larger models. ---8<-- -"snippets/enrollment-button.md" ---8<-- - --- diff --git a/latest/case_study/teaching/part03/README.md b/latest/case_study/teaching/part03/README.md index 6634406..bb6a415 100644 --- a/latest/case_study/teaching/part03/README.md +++ b/latest/case_study/teaching/part03/README.md @@ -884,8 +884,4 @@ If you achieve better than 55% v2 recall or discover interesting patterns, consi !!! note "The Improvement Flywheel in Action" This case study demonstrates the [improvement flywheel](../../../../docs/workshops/chapter0.md#the-improvement-flywheel-from-static-to-dynamic-systems) from Chapter 0 in practice. We started with synthetic evaluation data, identified the alignment problem through metrics, tested hypotheses systematically, and achieved a 358% improvement through iterative refinement. This is exactly how the product mindset transforms static RAG implementations into continuously improving systems. ---8<-- -"snippets/enrollment-button.md" ---8<-- - --- diff --git a/latest/case_study/teaching/part04/README.md b/latest/case_study/teaching/part04/README.md index fe69358..d8b425e 100644 --- a/latest/case_study/teaching/part04/README.md +++ b/latest/case_study/teaching/part04/README.md @@ -289,8 +289,3 @@ _This part demonstrates that reranking is a powerful technique for improving ret --- ---8<-- -"snippets/enrollment-button.md" ---8<-- - ---- diff --git a/latest/week0/README.md b/latest/week0/README.md index 50eceac..21cb82c 100644 --- a/latest/week0/README.md +++ b/latest/week0/README.md @@ -155,8 +155,4 @@ After completing this week's materials, you'll have: ## **Note**: This week's content is foundational. Take time to understand each concept thoroughly as they'll be used extensively throughout the course. ---8<-- -"snippets/enrollment-button.md" ---8<-- - --- diff --git a/latest/week1/README.md b/latest/week1/README.md index 7f58215..e8fe7bc 100644 --- a/latest/week1/README.md +++ b/latest/week1/README.md @@ -173,8 +173,3 @@ After completing this week's materials, you'll have: --- ---8<-- -"snippets/enrollment-button.md" ---8<-- - ---- diff --git a/latest/week2/README.md b/latest/week2/README.md index 8162d85..e24dd0a 100644 --- a/latest/week2/README.md +++ b/latest/week2/README.md @@ -162,8 +162,4 @@ After completing this week's materials, you'll have: ## **Note**: Ensure you've completed Week 1's evaluation framework before starting these notebooks. The metrics and benchmarking tools from Week 1 are essential for measuring the improvements achieved through fine-tuning. ---8<-- -"snippets/enrollment-button.md" ---8<-- - --- diff --git a/latest/week4/README.md b/latest/week4/README.md index 906d3f4..acd8ef7 100644 --- a/latest/week4/README.md +++ b/latest/week4/README.md @@ -161,8 +161,4 @@ After completing this week's materials, you'll have: ## **Note**: This week's analysis techniques are applicable beyond RAG systems. The same approach can help understand user behavior in any text-based application, making these skills broadly valuable for product improvement. ---8<-- -"snippets/enrollment-button.md" ---8<-- - --- diff --git a/latest/week5/README.md b/latest/week5/README.md index 1000a5a..a82773d 100644 --- a/latest/week5/README.md +++ b/latest/week5/README.md @@ -179,8 +179,4 @@ After completing this week's materials, you'll have: ## **Note**: The techniques in this week are essential for production RAG systems. Real-world queries often contain specific constraints that pure semantic search cannot handle, making structured data integration a critical capability. ---8<-- -"snippets/enrollment-button.md" ---8<-- - --- diff --git a/latest/week6/README.md b/latest/week6/README.md index fc82a7a..d614f4b 100644 --- a/latest/week6/README.md +++ b/latest/week6/README.md @@ -157,8 +157,4 @@ After completing this week's materials, you'll have: ## **Note**: The systematic approach demonstrated this week - metrics, testing, and targeted improvement - can be applied to any AI system optimization challenge. These skills transfer far beyond RAG applications. ---8<-- -"snippets/enrollment-button.md" ---8<-- - --- diff --git a/md/extra_kura_01_cluster_conversations.md b/md/extra_kura_01_cluster_conversations.md index d82ce29..b5f8751 100644 --- a/md/extra_kura_01_cluster_conversations.md +++ b/md/extra_kura_01_cluster_conversations.md @@ -555,8 +555,3 @@ By replacing vague summaries like "user seeks information about tracking" with p --- ---8<-- -"snippets/enrollment-button.md" ---8<-- - ---- diff --git a/md/extra_kura_02_better_summaries.md b/md/extra_kura_02_better_summaries.md index 2a8381a..606824f 100644 --- a/md/extra_kura_02_better_summaries.md +++ b/md/extra_kura_02_better_summaries.md @@ -483,8 +483,3 @@ This classifier will enable you to automatically categorize incoming queries in --- ---8<-- -"snippets/enrollment-button.md" ---8<-- - ---- diff --git a/md/extra_kura_03_classifiers.md b/md/extra_kura_03_classifiers.md index 41282ad..5898445 100644 --- a/md/extra_kura_03_classifiers.md +++ b/md/extra_kura_03_classifiers.md @@ -1519,8 +1519,3 @@ The key insight from this entire series is that improving RAG systems isn't abou --- ---8<-- -"snippets/enrollment-button.md" ---8<-- - ---- diff --git a/md/week6_03_improving_performance.md b/md/week6_03_improving_performance.md index 497f689..f9808d5 100644 --- a/md/week6_03_improving_performance.md +++ b/md/week6_03_improving_performance.md @@ -1058,8 +1058,3 @@ As we deploy these systems to production, this data-driven approach becomes even --- ---8<-- -"snippets/enrollment-button.md" ---8<-- - ---- diff --git a/md/week6_03_improving_performance_logfire.md b/md/week6_03_improving_performance_logfire.md index 34c6729..37a8b88 100644 --- a/md/week6_03_improving_performance_logfire.md +++ b/md/week6_03_improving_performance_logfire.md @@ -976,8 +976,3 @@ As we deploy these systems to production, this data-driven approach becomes even --- ---8<-- -"snippets/enrollment-button.md" ---8<-- - ----