How to summarize a long document
Get a useful summary without losing the thread — in about 5 minutes.
Open recipe →~5 min
Reference
How-to recipe library
W3Schools-style tasks — search or filter by time, then copy the steps.
54 recipes
How to write a system prompt
System prompts shape **every reply** in a session. Good ones are short, testable, and explicit about limits.
Open recipe →~5 min
How to chunk documents for RAG
Chunking makes or breaks RAG. Goal: each chunk is **small enough to retrieve** but **big enough to answer**.
Open recipe →~8 min
How to study a chapter with AI
Use AI as a **study partner**, not an answer key — especially for exams.
Open recipe →~10 min
How to build a simple chatbot
Minimal chatbot: send user text to an LLM API, show the reply. No RAG yet — that's the next step.
Open recipe →~15 min
How to debug a bad AI answer
Treat debugging like a checklist — don't blame the model until you know which stage failed.
Open recipe →~8 min
How to call OpenAI API with streaming
Streaming shows tokens as they arrive — better UX for long replies.
Open recipe →~10 min
How to add function calling
Let the model request actions; your code executes them safely.
Open recipe →~12 min
How to evaluate RAG quality
Measure retrieval and answers separately — a eloquent lie is still a failure.
Open recipe →~12 min
How to use AI for code review
Use AI as a **second pair of eyes** — not a merge approver.
Open recipe →~8 min
How to build a 5-document RAG app
Minimal RAG: five markdown/text files → embeddings → question → cited answer.
Open recipe →~20 min
How to write a few-shot prompt
Show the model **2–5 examples** of input → desired output so it copies the pattern.
Open recipe →~6 min
How to compare two models
Pick with **your tasks**, not leaderboard hype.
Open recipe →~10 min
How to summarize meeting notes
Turn messy notes into **decisions, actions, and open questions** — then verify names and dates.
Open recipe →~6 min
How to create an image with AI
Good images come from **specific prompts** — subject, style, lighting, composition.
Open recipe →~8 min
How to automate a workflow (no-code)
Chain triggers and AI steps without shipping production code — still review every output.
Open recipe →~10 min
How to pick an embedding model
Embeddings power RAG retrieval — pick for **language, domain, and cost**, not buzzwords.
Open recipe →~8 min
How to reduce hallucinations in answers
You can't eliminate hallucinations — but you can **ground, constrain, and verify**.
Open recipe →~8 min
How to set up a Python virtual environment
Isolate project packages so AI coding projects don't break your system Python.
Open recipe →~8 min
How to deploy a small AI app
Ship a minimal API or chat UI — secrets on the server, logs on, scope small.
Open recipe →~15 min
How to manage chat memory
Keep long chats useful without blowing the context window — in about 6 minutes.
Open recipe →~6 min
How to secure API keys
Keep provider keys off the client and out of Git — a 5-minute security pass.
Open recipe →~5 min
How to ship with a canary deploy
Send a small slice of traffic to the new model or prompt before full rollout.
Open recipe →~7 min
How to test prompt changes
Compare prompt v1 vs v2 on real tasks before you ship — about 6 minutes.
Open recipe →~6 min
How to wire Anthropic Messages API
Call Claude from Python with roles and a system prompt — first message in ~6 minutes.
Open recipe →~6 min
How to run a local LLM with Ollama
Get an offline chat running in about 8 minutes.
Open recipe →~8 min
How to handle API rate limits
Survive 429 errors without melting your budget.
Open recipe →~6 min
How to analyze a CSV with AI
One clear question, verified numbers — about 7 minutes.
Open recipe →~7 min
How to build a minimal agent loop
Think → tool → observe in code — starter pattern in ~10 minutes.
Open recipe →~10 min
How to run a canary deploy
Route 10% of traffic to the new stack — then promote or rollback.
Open recipe →~7 min
How to spot synthetic media
No single pixel test catches everything — combine **visual cues, context, and verification**.
Open recipe →~7 min
How to read an AI regulation summary
News articles oversimplify — use this to extract **what actually applies to you**.
Open recipe →~8 min
How to embed documents in batch
Index many chunks efficiently — **chunk once, embed in batches, store with metadata**.
Open recipe →~12 min
How to log production LLM calls
Debug quality, cost, and incidents without **leaking secrets or PII**.
Open recipe →~10 min
How to add citations to RAG answers
Users trust answers they can **click back to source chunks**.
Open recipe →~10 min
How to write a professional email with AI
Five minutes to a send-ready draft you still review line by line.
Open recipe →~5 min
How to compare reasoning vs fast models
Pick the right model tier with a tiny eval — not hype.
Open recipe →~7 min
How to deploy a RAG app (checklist)
Pre-flight before customer traffic hits retrieval + chat.
Open recipe →~8 min
How to wire an MCP server
Connect a tool server to your agent client in ~10 minutes.
Open recipe →~10 min
How to wire hybrid search for RAG
Keyword + vector + re-rank in a minimal pipeline.
Open recipe →~8 min
How to trace a failed RAG answer
Find the broken span before tuning prompts.
Open recipe →~6 min
How to ship an agent with an eval gate
No agent reaches prod without a golden task set.
Open recipe →~7 min
How to run golden-task eval locally
Score a prompt change before opening a PR.
Open recipe →~6 min
How to add a tool allowlist
Stop the model from calling tools you never intended to expose.
Open recipe →~5 min
How to design an AI workflow checkpoint
Human review before irreversible actions.
Open recipe →~5 min
How to red-team prompt injection
Automate adversarial prompts in staging before users find holes.
Open recipe →~6 min
How to train a LoRA adapter
Style and format without retraining the whole model.
Open recipe →~8 min
How to wire a multi-agent crew
Split work across an orchestrator and bounded workers.
Open recipe →~7 min
How to call a vision API
Image + question in one chat request.
Open recipe →~6 min
How to set production AI alerts
Watch latency, quality, and cost after you ship.
Open recipe →~5 min
How to profile token spend
Find where money goes before optimizing blindly.
Open recipe →~5 min
How to verify research bot answers
Citation gate before users see text.
Open recipe →~7 min
How to run an A/B model test
Compare two models or prompts on live traffic safely.
Open recipe →~6 min
How to triage an AI incident
1. Classify: data leak, quality regression, latency, or cost only
Open recipe →~6 min