LLM Zoomcamp: restructure for 2026 syllabus + Slack-derived FAQs

- Restructure modules to match the 2026 syllabus: module-2 is now Agents
  (was Vector Search), module-3 Vector Search (was Evaluation), module-4
  Evaluation (was Monitoring), module-5 Monitoring (new), module-6 Best
  Practices. Old workshops-agents folded into module-2 (Agents)
- Drop tooling that's no longer in the course: Qdrant content (not
  sponsoring this year), Elasticsearch troubleshooting (replaced),
  Arize/Phoenix (not used), Anaconda-specific install advice (uv is
  recommended now), 5 placeholder "Question" stubs, minsearch
  AppendableIndex (fixed in current minsearch)
- Add 12 technical FAQs from #course-llm-zoomcamp Slack threads
  (filtered to genuinely-technical-and-not-already-covered):
  - module-1: OpenAI 429 quota -> Groq/Gemini/Ollama; minsearch
    wget->raw syntax error; non-OpenAI token counting; direnv for
    API keys
  - module-2 (Agents): asyncio in Jupyter; non-OpenAI tool calling;
    OpenAI Responses API on non-OpenAI providers
  - module-3 (Vector Search): question vs answer vs both embedding
    strategy
  - module-4 (Evaluation): JSONDecodeError on ground truth; Jupyter
    kernel crash; Gemini rate limits
  - project: orchestration tool optional; recommender RAG eval;
    chunking strategy; reproducibility (no secrets in repo)
- Renumber sort_order sequentially within each section

ML Zoomcamp: drop Saturn Cloud content

- Remove 7 module-8 FAQs that were Saturn Cloud-specific (no longer
  sponsored). Update general/012 (special hardware) and misc/015
  (free cloud alternatives) to drop Saturn Cloud references.

Author: alexeygrigorev

_questions/llm-zoomcamp/_metadata.yaml

@@ -3,19 +3,27 @@ course_name: "LLM Zoomcamp"
 sections:
   - id: general
     name: "General Course-Related Questions"
+    comment: "Course logistics: cohort schedule, certificate, deadlines, leaderboard, project rules. Technical questions belong in the module sections."
   - id: module-1
-    name: "Module 1: Introduction"
+    name: "Module 1: Introduction to LLMs and RAG"
+    comment: "Basics of LLMs and RAG, OpenAI API, text search. Includes setup-related issues (API keys, ElasticSearch, Ollama, etc.)"
   - id: module-2
-    name: "Module 2: Vector Search"
+    name: "Module 2: Agents"
+    comment: "Adding agentic functionality to RAG, function calling, MCP, tool use"
   - id: module-3
-    name: "Module 3: Evaluation"
+    name: "Module 3: Vector Search"
+    comment: "Vector search, embeddings, indexing, Qdrant, semantic search"
   - id: module-4
-    name: "Module 4: Monitoring"
+    name: "Module 4: Evaluation"
+    comment: "Search evaluation, online vs offline evaluation, LLM as a judge, hit rate / MRR / ROUGE"
+  - id: module-5
+    name: "Module 5: Monitoring"
+    comment: "Online evaluation techniques, Phoenix, dashboards, monitoring user feedback"
   - id: module-6
-    name: "Module 6: X"
+    name: "Module 6: Best Practices"
+    comment: "Hybrid search, document reranking, advanced RAG techniques"
   - id: project
     name: "Capstone Project"
   - id: workshops-dlthub
-    name: "Workshops: dlthub"
-  - id: workshops-agents
-    name: "Workshops: Agents"
+    name: "Workshop: Open-Source Data Ingestion (dlt)"
+    comment: "Workshop on using dlt for ingesting data into RAG systems"

_questions/llm-zoomcamp/general/013_de32098d49_what-other-alternatives-to-elasticsearch-are-there.md

@@ -1,7 +1,7 @@
 ---
 id: 8d23fe792f
 question: 'Authentication: Safe and easy way to store and load API keys'
-sort_order: 12
+sort_order: 7
 ---
 
 You can store your different API keys in a YAML file that you will add to your `.gitignore` file. Be careful to never push or share this file.

_questions/llm-zoomcamp/module-1/007_66ae44829d_elasticsearch-error-badrequesterror-badrequesterro.md

@@ -1,7 +1,7 @@
 ---
 id: 233dabe430
 question: How to store and load API keys using .env file
-sort_order: 13
+sort_order: 8
 ---
 
 Store the API key in a `.env` file, then use the following steps to load it:

…n-safe-and-easy-way-to-store-and-load.md …n-safe-and-easy-way-to-store-and-load.md_questions/llm-zoomcamp/module-1/012_8d23fe792f_authentication-safe-and-easy-way-to-store-and-load.md renamed to _questions/llm-zoomcamp/module-1/007_8d23fe792f_authentication-safe-and-easy-way-to-store-and-load.md _questions/llm-zoomcamp/module-1/012_8d23fe792f_authentication-safe-and-easy-way-to-store-and-load.md renamed to _questions/llm-zoomcamp/module-1/007_8d23fe792f_authentication-safe-and-easy-way-to-store-and-load.md

@@ -1,7 +1,7 @@
 ---
 id: 86d99bbf21
 question: 'Authentication: Why is my OPENAI_API_KEY not found in the Jupyter notebook?'
-sort_order: 14
+sort_order: 9
 ---
 
 

…tore-and-load-api-keys-using-env-file.md …tore-and-load-api-keys-using-env-file.md_questions/llm-zoomcamp/module-1/013_233dabe430_how-to-store-and-load-api-keys-using-env-file.md renamed to _questions/llm-zoomcamp/module-1/008_233dabe430_how-to-store-and-load-api-keys-using-env-file.md _questions/llm-zoomcamp/module-1/013_233dabe430_how-to-store-and-load-api-keys-using-env-file.md renamed to _questions/llm-zoomcamp/module-1/008_233dabe430_how-to-store-and-load-api-keys-using-env-file.md

@@ -3,7 +3,7 @@ id: fe8fed31e6
 question: 'OpenSource: I am using Groq, and it doesn''t provide a tokenizer library
   based on my research. How can we estimate the number of OpenAI tokens asked in homework
   question 6?'
-sort_order: 15
+sort_order: 10
 ---
 
 The question asks for the number of tokens in the GPT-4o model. `tiktoken` is a Python library that can be used to get the number of tokens. You don't need an OpenAI API key to get the number of tokens. You can use the code provided in the question to get the number of tokens.