Catalogue

Learning resources from Anthropic, OpenAI, Google, Perplexity, Hugging Face, xAI and independent writers. Every summary is written fresh, and every entry links to the original. Where we don't know something (a price, a licence), it says unknown rather than guessing.

All 97 resources: the six AI providers first, then independent sources

Resource	Provider	Type	Audience	Cost	Login	Licence	Last checked
AI Capabilities and Limitations Garden A plain-language course on how AI chat tools work: why they predict text, why their knowledge stops at a certain date, how much they can hold in mind at once, and how to figure out what went wrong when an answer surprises you. It is for anyone who uses AI at work or school and wants to judge its output with more confidence. Explains: Large language model, Context window, Hallucination, Jagged frontier, Verification	Anthropic	course	Nontechnical	free	Optional	unknown	2026-10-16
AI Fluency for Nonprofits Garden Applies the AI Fluency habits to nonprofit work such as fundraising, communications, programs and operations. You practise research, writing and data tasks, think through privacy and data handling, and work toward an AI policy for your organisation. Best taken after the Framework & Foundations course. Explains: Prompting, Privacy and data, Verification, Automation	Anthropic	course	Nontechnical	free	Optional	unknown	unknown
AI Fluency: Framework & Foundations Garden A free self-paced course that gives you a simple way to think about working with AI: deciding what to hand over, explaining the job clearly, judging what comes back, and taking responsibility for how you use it. It suits anyone starting out, with no technical background needed, and ends with a short quiz and a badge. Explains: Generative AI, Prompting, Verification, Human in the loop, Automation	Anthropic	course	Nontechnical	free	Optional	CC BY-NC-SA 4.0	2026-10-16
Anthropic Education Report: How university students use Claude Garden A look at how university students use Claude. Nearly half of conversations were quick answer-seeking, and the AI was most often doing the higher-level thinking, such as creating and analysing, rather than the student. It asks what students lose when they hand the hard parts to the AI. Explains: Cognitive offloading, Productive friction	Anthropic	research report	Nontechnical	free	No login	unknown	2026-10-16
Claude 101 Garden A guided tour of Claude for people starting to use it at work. It walks through everyday prompting, then features such as Projects, artifacts, Skills, connectors to other apps and research, with examples by job role. Explains: Prompting, Projects and knowledge, Artifacts and canvases, Skills, Tools and connectors	Anthropic	course	Nontechnical	free course; some hands-on features need a paid Claude plan	Optional	unknown	unknown
Claude Help Center: Introduction to prompt design Garden A short help article on writing requests to Claude. Its main point is that Claude knows nothing about you or your task, so you should brief it the way you would brief a new contractor, with clear and specific instructions. Explains: Prompting, Prompt engineering, Context engineering	Anthropic	help article	Mixed	free	No login	all rights reserved	unknown
How AI assistance impacts the formation of coding skills Garden A controlled experiment in which people learned a new programming tool with or without AI help. Those who leaned on AI scored lower on a follow-up quiz, but people who used AI to ask questions and understand ideas kept their learning and still worked fast. The finding applies beyond coding: the way you use AI affects how much you learn. Explains: Cognitive offloading, Productive friction, Human in the loop	Anthropic	research study	Mixed	free	No login	unknown	2026-10-16
Introduction to Claude Cowork Garden Shows people who shuffle information between files, apps and tools how to hand multi-step work to Claude's Cowork mode: setting up a workspace, giving it context, and running tasks on a schedule. No coding needed, but you need a paid Claude plan to follow along. Explains: Agents, Scheduled tasks, Skills, Context engineering, Automation	Anthropic	course	Nontechnical	free course; Cowork itself needs a paid Claude plan	Optional	unknown	unknown
The AI Fluency Index Garden Anthropic studied about ten thousand real Claude conversations to see which good habits people show. People who kept refining their request were far more likely to question the AI's reasoning, while polished outputs got less scrutiny, and very few people checked facts. Read it for the evidence behind two habits: keep refining your request, and check polished output as closely as rough drafts. Explains: Verification, Steering, Artifacts and canvases, Productive friction	Anthropic	research report	Nontechnical	free	No login	unknown	2026-10-16
Gemini API prompt design strategies Garden Google's guide to designing prompts for Gemini, written for developers but readable by confident nontechnical users. It covers giving clear instructions, adding examples, supplying context, breaking a request into parts and iterating on a prompt that isn't working. Explains: Prompting, Prompt engineering, Examples in prompts, Context engineering	Google	developer guide	developer	free	No login	CC BY 4.0 (text); Apache-2.0 (code samples)	2026-10-16
Gemini for Workspace prompting guide (PDF) Garden A downloadable guide from Google Workspace on writing prompts for Gemini inside everyday office tools. It is aimed at knowledge workers who use Gmail, Docs and similar apps. Explains: Prompting, Prompt engineering	Google	guide	Nontechnical	free	No login	all rights reserved	unknown
Google AI Essentials Garden A short certificate course from Google, delivered through Coursera, for people with no AI experience. It runs across five modules in under five hours. Note that it is paid after a free trial. Explains: Generative AI, Prompting	Google	certificate course	Nontechnical	paid: $49/month on Coursera after a 7-day free trial (US/Canada)	Required	all rights reserved	2026-10-16
Google Workspace Learning Center Garden Google's help hub for Workspace apps, including how-to material on using Gemini in Docs, Gmail and other tools. Use it to look up a specific task rather than to learn concepts. Explains: Tools and connectors	Google	help hub	Nontechnical	free	Unknown	all rights reserved	unknown
Introduction to Generative AI Garden A 45-minute introductory course explaining what generative AI is, what it is used for, and how it differs from older kinds of machine learning. It ends with a badge and needs no background. Explains: Generative AI, Large language model, Multimodal AI	Google	microlearning course	Nontechnical	free (UNVERIFIED)	Optional	all rights reserved	2026-10-16
Agents Course, Unit 0-1 (what an agent is) Garden The first units of Hugging Face's free agents course explain what an AI agent is: a model that can plan steps and use tools rather than only answer. The early units are conceptual; the rest of the course is aimed at developers. Explains: Agents, Tools and connectors, Large language model	Hugging Face	course	Mixed	free	Optional	Apache-2.0	2026-10-16
HuggingChat (try open models) Garden A chat app where you can talk to many open models, either letting it pick one for you or choosing a model yourself. Use it for practice: try the same task on different models and compare what you get. Explains: Model choice, Large language model	Hugging Face	tool	Nontechnical	free tier	Required	Hugging Face terms of service (proprietary site content)	2026-10-16
LLM Course, Chapter 1: Transformer models (intro) Garden The opening chapter of Hugging Face's free course on large language models, which explains what these models are and what they can do. Curious nontechnical readers can follow this first chapter; later chapters need Python. Explains: Large language model, Generative AI, Model choice	Hugging Face	course chapter	Mixed	free	No login	Apache-2.0	2026-10-16
ChatGPT 101: A Guide to Your AI Super Assistant (recording) Garden A recorded introductory session from OpenAI Academy for people new to ChatGPT. It is a starting point for getting comfortable with the tool before moving on to role-specific material. Explains: Generative AI, Prompting	OpenAI	recorded webinar	Nontechnical	free	Unknown	all rights reserved	unknown
OpenAI Academy: AI Foundations Garden OpenAI's beginner course for people who use AI at work, which runs inside ChatGPT and ends with a certificate. Details of its contents and access come from secondary sources only, so treat this entry as a pointer until someone checks the course page. Explains: Generative AI, Prompting	OpenAI	course	Nontechnical	free (secondary sources only)	Unknown	all rights reserved	unknown
OpenAI Academy: Apply AI at Work track Garden A hub page collecting OpenAI Academy courses and events on using AI in everyday work, with separate groups for small businesses, nonprofits and government. Use it to find a role-based next step after a beginner course. Explains: Prompting, Automation	OpenAI	course track	Nontechnical	free	Unknown	all rights reserved	2026-10-16
OpenAI Academy: Nonprofits club Garden An OpenAI Academy community space for nonprofit staff, with dated events and sessions. Useful if you want to learn alongside peers in the sector rather than on your own. Explains: Generative AI	OpenAI	community hub	Nontechnical	free	Required	all rights reserved	unknown
Prompt engineering best practices for ChatGPT Garden OpenAI's help-center article on writing better requests to ChatGPT. It is the ChatGPT counterpart to the other providers' prompting guides; its contents have not been checked yet. Explains: Prompting, Prompt engineering	OpenAI	help article	Nontechnical	free	No login	all rights reserved	unknown
Evaluating AI for high-stakes work Garden A guide for teams in finance, legal and research who must check AI work before it shapes a decision. It reviews output for accuracy, transparency and human oversight, and helps you spot risky workflows and choose what to automate first. Explains: Verification, Human in the loop, Hallucination, Evals	Perplexity	guide	Mixed	free (gating UNVERIFIED)	Unknown	all rights reserved	unknown
Perplexity Academy Garden A library of short video lessons, a few minutes each, on using Perplexity's search and its Computer feature for multi-step work. Most lessons are product how-tos rather than general skills. Explains: Deep research, Agents	Perplexity	video lessons	Nontechnical	free to watch; Computer features need a paid or Enterprise plan (UNVERIFIED)	Unknown	all rights reserved	2026-10-16
Perplexity at Work Guide Garden A downloadable guide on using Perplexity in daily work, organised around cutting distractions, extending what you can research and write, and reaching concrete outcomes. It reads as product marketing, so use it for ideas rather than method. Explains: Deep research, Automation	Perplexity	guide	Nontechnical	free; may be gated by a form (UNVERIFIED)	Unknown	all rights reserved	unknown
Search vs Computer Garden A short lesson on when a quick cited answer is enough and when to hand over a multi-step job that pulls from several tools, builds a report and can run on a schedule. It ends with a simple checklist for choosing between the two. Explains: Model choice, Deep research, Agents, Scheduled tasks	Perplexity	video lesson	Nontechnical	free to watch	Unknown	all rights reserved	unknown
Grok Bot use cases Garden A page of example roles for xAI's Grok Bot, which gathers and prepares work across connected tools and hands it back for review, with outside actions held for your approval. Examples include sales outreach, recruiting, expenses and a chief-of-staff assistant. Use it for project ideas; it does not teach skills. Explains: Agents, Tools and connectors, Human in the loop, Scheduled tasks	xAI	use-case page	Nontechnical	unknown; the product is likely paid (UNVERIFIED)	Unknown	all rights reserved	unknown
Welcome to Grok (consumer docs) Garden xAI's overview of the Grok assistant and its features. It describes the product rather than teaching skills, so it is mainly useful when you want to try the same task in Grok as in other assistants. Explains: Model choice	xAI	product docs	Nontechnical	free tier; SuperGrok is paid	Required	all rights reserved	2026-10-16
AI Should Challenge, Not Obey Garden Advait Sarkar argues that AI should sometimes act as a provocateur, offering counterarguments, critiques and questions instead of finishing your work for you. He also warns that constant criticism gets tiresome, so the challenge has to fit the moment. Short and readable for anyone curious about AI as a tool for thinking. Explains: Productive friction, Sycophancy, Cognitive offloading	ACM	essay	Mixed	free	No login	unknown	2026-10-16
The Impact of Generative AI on Critical Thinking Garden A survey of 319 knowledge workers about when they think critically while using AI. People who trusted the AI more tended to think less critically, and people more confident in their own skills tended to think more. Their critical effort moved toward checking, combining and overseeing AI work. Explains: Cognitive offloading, Verification, Human in the loop	CHI	paper	Mixed	free (open access, per research notes)	No login	CC BY	unknown
Claude Cowork 101: How to automate your workday without touching code (How I AI, with JJ Englert) Garden JJ Englert shows how someone who doesn't code can set up Claude Cowork: a "brain" file that tells Claude about you and your work, an email-writing skill built from your own sent mail, a panel of simulated reviewers that critique one draft from several angles, and a read-only morning plan. Good for knowledge workers who already use a chat assistant and want a first repeatable routine. Explains: Skills, Projects and knowledge, Custom instructions, Scheduled tasks, Tools and connectors	Claire Vo	podcast episode	Nontechnical	free to watch; Claude Cowork and the connected tools have their own plans (not checked)	No login	unknown	2026-10-16
Claude Cowork for PMs: My Self-Improving Productivity System (How I AI, with Daniel Blum) Garden Daniel Blum, a product manager at Melio, shows a Claude Cowork setup that preps his week, writes a morning brief from meeting notes and messages, and asks him to define internal terms it doesn't know before saving them. A weekly pass compares its drafts with his edits and suggests new skills. Watch it to see an assistant improve through many small confirmed corrections instead of one big setup. Explains: Memory, Context engineering, Scheduled tasks, Skills, Human in the loop, Compounding loop	Claire Vo	podcast episode	Mixed	free to watch; tools shown (Claude Cowork, Notion, Granola) have their own plans	No login	unknown	2026-10-16
How I AI (podcast) Garden A weekly video show hosted by Claire Vo where guests share their screen and show how they use AI in their actual job: product managers, designers, founders, engineers and parents. Because you watch the work happen, it is easier to copy than an interview. For a first episode, pick one aimed at people who don't code, such as the Claude Cowork tutorial with JJ Englert. Explains: Agents, Skills, Automation, Vibe coding	Claire Vo	podcast	Mixed	free	No login	unknown	2026-10-16
How I AI workflow guides Garden About 290 short step-by-step guides, each taken from a How I AI episode where a guest shows a real AI workflow on screen. You can filter by tool, job function (operations, research, content and more) and difficulty, so a beginner can find something small to try this week. Each guide lists the tools, the inputs, the steps and how to check the result. Explains: Skills, Automation, Agents, Custom assistants	Claire Vo	workflow library	Mixed	free	No login	unknown (pages say © 2026 ChatPRD, Inc.; downloadable skill files name no licence)	2026-10-16
How to Improve Interview Feedback Consistency with a Custom GPT Garden A beginner guide to a custom GPT that holds your interview rubric plus strong and weak example scorecards, and reviews how well interviewers wrote up their evidence. The AI is kept to a narrow job: it rates the write-up, never the candidate, makes no hire decision, and flags claims tied to personal characteristics. A reviewer edits its note before sending. A good model for giving AI one bounded job with a rubric. Explains: Custom assistants, Examples in prompts, Evals, Privacy and data, Human in the loop	Claire Vo	workflow guide	Nontechnical	free to read; a custom GPT needs a ChatGPT plan that supports them (not checked)	No login	unknown (© 2026 ChatPRD, Inc.)	2026-10-16
How to Turn a One-Off AI Chat Session into a Reusable Workflow for Your Team Garden Taken from the How I AI episode on Kai, Stripe's internal AI assistant. After a chat that produced something you want again, such as a dashboard, you ask the AI to package the session as a skill, edit its name and trigger phrases, add it to a shared library, then test it in a fresh session. Kai is internal to Stripe; you would need to try the same steps in an assistant that supports skills. Explains: Skills, Verification, Compounding loop	Claire Vo	workflow guide	Mixed	free	No login	unknown (© 2026 ChatPRD, Inc.)	2026-10-16
How to Write a PRD: A Step-by-Step Guide for Product Managers Garden A ten-step guide to writing a product requirements document: the problem, who it is for, goals, user stories, priorities, risks and how to get people to agree, with a good and a bad example and three free templates. It is written for product managers, but anyone who has to describe what should be built before asking a person or an AI to build it can use the problem and goal steps. Explains: Context engineering, Smallest useful version	Claire Vo	guide	Mixed	free	No login	unknown (© 2026 ChatPRD, Inc.)	2026-10-16
OpenClaw: The complete guide to building, training, and living with your personal AI agent Garden Claire Vo's guest post on setting up OpenClaw, an open-source personal agent you message through apps such as Telegram. She explains why it should run on a separate machine with its own accounts, how its identity and rules live in a few text files, and six starter jobs such as pre-meeting briefs and weekend logistics. The setup part is free to read; the section on running several agents is for paid subscribers. Explains: Agents, Scheduled tasks, Skills, Privacy and data, Tools and connectors, Onboard AI like a new hire, Prompt injection	Claire Vo	article	Mixed	paid (Lenny's Newsletter subscription) for the full post; first part free	Required	unknown	2026-10-16
The AI Agent Glossary Garden A community glossary, started from a post by Claire Vo, of agent words such as agent, sandbox, context window, system prompt and MCP. Each term has a plain definition and an "in practice" example. It is one Markdown file, so you can hand it straight to your own AI. Some examples quote early-2026 prices and older models, so check numbers before you repeat them. Explains: Agents, Model Context Protocol (MCP), Context window, Tools and connectors, Large language model	Claire Vo	glossary	Nontechnical	free	No login	CC0-1.0 (declared in the README; the repo has no LICENSE file)	2026-10-16
Tradclaw: a household starter kit for OpenClaw Garden A starter set of files for running a personal AI agent as a family manager: calendar briefs, school email triage, meal plans and shopping lists, home maintenance reminders and stories for kids. The agent interviews you first and switches on only the modules you choose. Worth reading for its safety rules even if you never install it: it asks before sharing anything about children, and it never treats text from email, web pages or calendar invites as an instruction. Explains: Agents, Scheduled tasks, Privacy and data, Memory, Skills, Onboard AI like a new hire, Prompt injection	Claire Vo	repository	Mixed	free; model and hosting costs depend on your setup	No login	none (GitHub reports no licence)	2026-10-16
Why "Get With the Program" Doesn't Work: Claire Vo on AI That Actually Sticks (Agents of Scale) Garden A 51-minute conversation on Zapier's Agents of Scale podcast about why AI mandates and hiring freezes don't get a team using AI, and what does: regular rituals such as AI Fridays and hack weeks, leaders who build things themselves in the open, and treating AI skill as career growth rather than compliance. For anyone who has been asked to "get the team using AI". Explains: Vibe coding, Adoption rituals over mandates	Claire Vo	podcast episode	Nontechnical	free	No login	unknown	2026-10-16
AI Style Guides: How to Help AI Write Like You Garden A step-by-step guide to writing a style guide that an AI reads before it drafts for you. You gather a few writing samples, let the AI interview you about your voice and what you dislike in AI prose, react to sample lines, and turn the answers into a guide you test and revise over a week. Useful for anyone who keeps rewriting AI drafts to sound like themselves. Explains: Custom instructions, Examples in prompts, Clarifying questions, AI writing tells, Projects and knowledge	Every	guide	Nontechnical	free	No login	unknown	2026-10-16
An Executive's Guide to Implementing AI Garden A five-step plan for leaders bringing AI into an organisation: use the tools yourself first, give named champions protected time, pick one frequent and painful workflow, build it with good examples, checks and a named owner, then spread what works and drop what fails. It argues for maintained, owned skills libraries over loose prompt collections. For leaders and the operators who run adoption day to day. Explains: Skills, Evals, Human in the loop, Smallest useful version, Adoption rituals over mandates	Every	guide	Nontechnical	free	Required	unknown	2026-10-16
ChatGPT for Knowledge Work Garden A guide to using ChatGPT's agent features for office work, split into tasks you can hand off and tasks you should work through together. It moves through five levels from one-off tasks to recurring workflows and ends with a library of sixteen worked workflows, such as a weekly KPI report, a research brief and a recruiting shortlist, plus a seven-day starter plan. For managers and small teams who already use ChatGPT. Explains: Agents, Tools and connectors, Scheduled tasks, Skills, Automation	Every	guide	Nontechnical	paid Every subscription (price not checked)	Required	unknown	2026-10-16
Compound Engineering Garden Every's long guide to working with AI agents so that each job leaves the next one easier. It walks through a plan, work, review and "compound" loop, where the last step writes down what you learned so the agent catches the same problem next time, and a six-stage ladder from doing everything by hand to running several agents at once. Written for software teams, but the loop and the ladder carry over to any repeated knowledge work. Explains: Agents, Skills, Custom instructions, Steering, Human in the loop, Context engineering, Compounding loop	Every	guide	developer	free	Optional	unknown	2026-10-16
Compound Writing: The Ultimate Guide Garden Applies the compound loop to writing: interview, outline, draft, review, finish, then save what you learned so the next piece goes faster. It keeps your voice and project rules in plain files the agent reads each time. For people who write at work and are willing to run an agent like Claude Code or Codex. Explains: Custom instructions, Skills, Projects and knowledge, Agents, Compounding loop	Every	guide	Mixed	paid Every subscription (price not checked)	Required	unknown	2026-10-16
How to Use Claude Code as a Second Brain Garden An AI & I episode where Noah Brier shows how he points Claude Code at his folder of notes and tells it he is in "thinking mode", so it asks him questions and finds his past research instead of drafting. He keeps a project folder with saved chats, a daily progress log and sources, and asks the agent to recap recent days when he picks the work back up. For people who think through writing and keep notes. Explains: Clarifying questions, Agents, Projects and knowledge, Cognitive offloading, Productive friction	Every	podcast episode	Mixed	unknown	Unknown	unknown	2026-10-16
Securing an Always-on AI Employee Garden How Every limits the risk of an agent that runs all day with real accounts. It covers three threats (poisoned software packages, instructions hidden in content the agent reads, and leaks of internal data) and four layers of defence: give the agent its own accounts and only the access it needs, add hard blocks it cannot talk its way past, write rules into its instructions, and log everything. Best for whoever decides how much access a team's agent gets. Explains: Agents, Privacy and data, Human in the loop, Tools and connectors, Onboard AI like a new hire, Prompt injection	Every	guide	Mixed	free	No login	unknown	2026-10-16
The Agent That Saved My Brain Garden Every's head of growth, who says he has no technical background, describes the "command center" he built in Claude Code: a set of notes about the business, connections to the tools he uses, and dozens of small skills for jobs he repeats, such as turning a voice note into a campaign brief. He released part of it as the Compound Knowledge plugin. A case study of a non-engineer growing a personal agent one task at a time. Explains: Agents, Skills, Tools and connectors, Projects and knowledge, Model Context Protocol (MCP), Compounding loop	Every	essay	Nontechnical	unknown	Unknown	unknown	2026-10-16
The Eight Levels of AI Adoption Garden Lays out eight levels of handing work to AI, from a chatbot that answers one request, through an agent that checks in before each step and one that works without approvals, up to a manager agent directing others. It says most knowledge workers get the most from the first four levels and that a higher level is not automatically better. Helps you name where you are and what the next step up would ask of you. Explains: Agents, Human in the loop, Automation, Solution ladder, Steering, Delegating to AI	Every	guide	Nontechnical	free	Required	unknown	2026-10-16
The Knowledge Economy Is Over. Welcome to the Allocation Economy. Garden Dan Shipper argues that as AI takes over more of the making, people get paid less for what they know and more for how well they direct models. He names four skills a "model manager" needs: a clear goal, the taste to say exactly what is wrong with a result, judging which model suits which job, and knowing when to get into the details and when to leave the work alone. A short, readable framing for anyone starting to delegate to AI. Explains: Steering, Model choice, Human in the loop, Allocation economy	Every	essay	Nontechnical	free	No login	unknown	2026-10-16
We Gave Every Employee an AI Agent. Here's What Happened. Garden Every's COO and head of platform talk through giving each staff member their own agent. Lessons include starting with personal chores so the value is obvious, running agents in shared channels so people learn from each other and owners stay accountable, and expecting agents in group chats to pile on or loop. They treat it as a culture change, because people differ a lot in how well they delegate. Explains: Agents, Memory, Human in the loop, Skills	Every	podcast transcript	Nontechnical	free	No login	unknown	2026-10-16
How to be more ambitious: A tutorial for humans with models Garden A practical essay on asking more of AI. It describes kinds of ambition, such as aiming at the goal behind your goal, asking for a deeper version of a task rather than a bigger one, and having the AI play the real roles of a team. It also advises against letting a model judge its own work. Explains: Prompting, Evals, Verification, Jagged frontier	Felix Rieseberg	essay	Nontechnical	free	No login	unknown	2026-10-16
A guide to AI prototyping for product managers Garden Colin Matthews compares three kinds of tool for turning an idea into a clickable prototype: chat assistants, browser builders such as Bolt, v0, Lovable and Replit, and coding assistants such as Cursor, and says when to use each. He gives prompt templates for common jobs, like turning a sketch or a requirements doc into a prototype, and ways to get unstuck without coding: ask it to reflect, change one thing at a time, be specific, and start fresh when it loses track. Explains: Vibe coding, Artifacts and canvases, Solution ladder	Lenny	article	Mixed	paid on lennysnewsletter.com; full text free in Lenny's public data starter pack (personal, non-commercial use)	Required	all rights reserved (Lenny Rachitsky)	2026-10-16
An AI glossary Garden Plain-language definitions of about 20 AI terms, from model, token and training to RAG, MCP, agents, evals and hallucination, written for people in meetings where these words fly past. Lenny says he adds terms as new ones appear. If you need text you can reuse, use the open AI Agent Glossary instead. Explains: Large language model, Generative AI, Hallucination, Model Context Protocol (MCP), Agents, Evals	Lenny	glossary	Nontechnical	paid on lennysnewsletter.com; full text free in Lenny's public data starter pack (personal, non-commercial use)	Required	all rights reserved (Lenny Rachitsky)	2026-10-16
Beyond vibe checks: A PM's complete guide to evals Garden Aman Khan of Arize AI explains how to test AI output on purpose rather than by feel. He covers what to measure (made-up facts, relevance, tone), a four-part recipe for a prompt that asks one AI to grade another's answer (a role, the text to grade, the goal, and exact definitions of each label), and a cycle of collecting examples, grading them, fixing the prompt and monitoring. The recipe works for anyone who reuses a prompt, not only product teams. Explains: Evals, Verification, Hallucination	Lenny	article	Mixed	paid on lennysnewsletter.com; full text free in Lenny's public data starter pack (personal, non-commercial use)	Required	all rights reserved (Lenny Rachitsky)	2026-10-16
Build your personal AI copilot Garden Tal Raviv's free guide to using an AI project as a long-term thinking partner at work. You "hire" it by writing how it should work with you (including when to push back), onboard it with background on your role, team and goals, kick off one initiative, then keep feeding it updates so its advice gets better. It includes starter prompts and an appendix for tools without a projects feature. Explains: Projects and knowledge, Custom instructions, Context engineering, Memory, Sycophancy, Onboard AI like a new hire	Lenny	article	Nontechnical	free	No login	unknown	2026-10-16
Everyone should be using Claude Code more Garden Lenny walks through installing Claude Code and then collects 50 ways people outside engineering use it: cleaning up messy invoice files, organising folders, synthesising customer-call transcripts, drafting a job description, building slides. Read it for ideas once chat feels comfortable and you want an AI that works on the files on your own computer. Explains: Agents, Vibe coding, Tools and connectors	Lenny	article	Mixed	paid on lennysnewsletter.com; full text free in Lenny's public data starter pack (personal, non-commercial use)	Required	all rights reserved (Lenny Rachitsky)	2026-10-16
How to build your PM second brain with ChatGPT Garden Amir Klein, a product manager at monday.com, shows how he set up a ChatGPT Project as a second brain for a new initiative: write instructions that make it challenge him, upload everything he had (docs, decks, exported Slack threads) as files, then ask it to pull together plans and gaps. An appendix covers Claude and Gemini. It is aimed at product managers, but the steps fit anyone juggling scattered context. Explains: Projects and knowledge, Custom instructions, Context engineering, Sycophancy, Onboard AI like a new hire	Lenny	article	Mixed	paid on lennysnewsletter.com; full text free in Lenny's public data starter pack (personal, non-commercial use)	Required	all rights reserved (Lenny Rachitsky)	2026-10-16
Lenny's Data: free public starter pack Garden A free sample of Lenny's archive as plain Markdown: 10 newsletter posts and 50 podcast transcripts, with an index of titles, dates and guests. It includes several posts that are otherwise paid, among them the AI prototyping, evals, AI glossary and second-brain guides. Use it to read those posts, or to practise asking your own AI questions over a folder of documents. Explains: Projects and knowledge, Context engineering, Model Context Protocol (MCP)	Lenny	dataset	Mixed	free starter pack; full archive (373 posts, 321 transcripts, MCP access) for paid subscribers	No login	Lenny's Data starter-pack terms: personal, non-commercial use; no redistribution of raw files; © Lenny Rachitsky, all rights reserved	2026-10-16
Not all AI agents are created equal Garden Sorts agent ideas into three kinds before you compare effort or value: fixed step-by-step automations where AI fills in content at set points, agents that choose their own next step from a set of tools, and networks of agents owned by different teams. The authors advise starting with the first kind for the quickest measurable return. Most of the post, including the third kind, is for paid subscribers. Explains: Agents, Automation, Solution ladder	Lenny	article	Mixed	paid (Lenny's Newsletter subscription); opening sections free	Required	unknown	2026-10-16
The non-technical PM's guide to building with Cursor (Lenny's Podcast, with Zevi Arnovitz) Garden Zevi Arnovitz, a product manager at Meta who doesn't write code, shows how he builds working products with Cursor and Claude. Saved commands take him from capturing an idea as an issue, to exploring the problem, writing a plan, building, and asking several different AI models to review the result. He suggests starting in a simple chat project and moving to heavier tools only when you outgrow them. Explains: Vibe coding, Steering, Skills, Solution ladder, Verification	Lenny	podcast episode	Mixed	free to watch; the tools shown have their own plans	No login	unknown	2026-10-16
This week on How I AI: Claude Cowork tutorial for non-engineers + Build your own Slack inbox (for $0) Garden A free weekly roundup of two How I AI episodes: JJ Englert's Claude Cowork tutorial for people who don't code, and Yash Tekriwal's custom Slack digest, which cut 100 to 150 daily notifications to about 30 that need attention. Each comes with links to step-by-step workflow write-ups. A quick way to decide whether either episode is worth an hour. Explains: Automation, Skills, Scheduled tasks	Lenny	newsletter roundup	Nontechnical	free	No login	unknown	2026-10-16
What people are vibe coding (and actually using) Garden More than 50 small tools that readers built with Replit, Lovable, Bolt, Cursor and similar and still use: a carb counter, a chore app for kids, a bill splitter, meeting-prep automation, a tracker of wins at work. Each entry names the builder and the tool. Useful for choosing a first build that is small, personal and likely to get used. Explains: Vibe coding, Smallest useful version, Artifacts and canvases	Lenny	article	Nontechnical	paid on lennysnewsletter.com; full text free in Lenny's public data starter pack (personal, non-commercial use)	Required	all rights reserved (Lenny Rachitsky)	2026-10-16
A Conversation with Dr. Vivienne Ming, author of Robot-Proof Garden A short written Q&A from Fidelity's research centre. Ming explains her split between well-posed problems that language models handle well and open problems where people do better, and why the combination beats either alone. Her advice for individuals: map the known part with AI, do your own reasoning on the uncertain part, and question its answers. For leaders: develop curious, humble, adaptable people, since she finds those traits predict who works well with AI. Explains: Productive friction, Cognitive offloading, Human in the loop, Centaurs and cyborgs	Ming	interview	Nontechnical	free	No login	unknown	2026-10-16
A top researcher says a new divide is emerging in AI use, and most people are on the losing side Garden A news report on Ming's forecasting experiment, in which teams predicted real events with or without AI and were compared against Polymarket. She reports that most people either let AI produce the answer or used it to confirm what they already believed. A small minority, roughly 5-10%, argued with it and asked it to explain why they were wrong, and those people did best. The finding is reported in an interview and has not been published as research. Explains: Productive friction, Sycophancy, Cognitive offloading, Centaurs and cyborgs	Ming	article	Nontechnical	free	No login	unknown	2026-10-16
In Conversation... with Vivienne Ming Garden A 34-minute interview from a law firm's future-of-work series, with a written transcript. Ming describes three ways people use AI: self-automators who hand over the thinking, centaurs who split the work, and cyborgs who actively direct and argue with it. She says only the third makes people better at their jobs. Her practical advice: ask AI to critique your draft and then fix it yourself, and protect the routine tasks junior staff learn from. Explains: Productive friction, Cognitive offloading, Human in the loop, Centaurs and cyborgs	Ming	podcast	Nontechnical	free	No login	unknown	2026-10-16
Robot-Proof: When Machines Have All the Answers, Build Better People Garden Vivienne Ming's 2026 book about people more than machines. She argues that AI handles questions with known answers well, so human value lies in open, ill-posed problems, and in traits such as curiosity, humility and perspective-taking that can be developed. It covers what predicts long-term outcomes for children, what makes teams smart, and why using AI to skip thinking may cost us. For readers who want the case for building people alongside tools. Explains: Cognitive offloading, Productive friction, Human in the loop	Ming	book	Nontechnical	paid: hardcover $30.00, ebook $18.00 (Wiley list prices)	No login	all rights reserved (commercial book)	2026-10-16
The AI That Asks Better Questions: Dr. Vivienne Ming at SIM 2025 Garden A write-up of Ming's 2025 keynote in Porto. Her line is that AI should make your life harder in ways that make you better. She describes giving her book manuscript to a language model and asking it to find every place she lied, misled or oversimplified, and she says the best AI tutors hold back answers and ask questions. A quick introduction to her case for AI as a critic. Explains: Productive friction, Verification, Cognitive offloading	Ming	article	Nontechnical	free	No login	unknown	2026-10-16
Vivienne Ming on hybrid collective intelligence, building cyborgs, meta-uncertainty, and the unknown infinite Garden A long interview on the Humans + AI podcast with a full transcript. Ming's test for any tool is whether it leaves you more capable once you switch it off. She suggests pausing to judge your own uncertainty before accepting an answer, using AI to critique your thinking, and treating it like a research assistant you challenge and correct. She also explains why diverse teams produce more novel ideas and why AI tends to make people's ideas more alike. Explains: Productive friction, Cognitive offloading, Verification, Sycophancy, Centaurs and cyborgs	Ming	podcast	Nontechnical	free	No login	unknown	2026-10-16
Your Brain on ChatGPT Garden An MIT Media Lab study that measured brain activity while people wrote essays with ChatGPT, a search engine or nothing. The ChatGPT group felt least ownership and often could not quote their own essays. People who wrote alone first and then used AI stayed more engaged, which supports thinking before you prompt. Explains: Cognitive offloading, Productive friction, AI writing tells	MIT Media Lab	paper	Mixed	free	No login	unknown	unknown
15 Times to use AI, and 5 Not to Garden A checklist for deciding whether a task suits AI. Good fits include generating many ideas, work in your own field that you can judge quickly, low-stakes summaries and getting unstuck on a small problem. Poor fits include learning something you need to understand yourself, work that needs very high accuracy, and tasks where the struggle is how you reach the insight. Explains: Jagged frontier, Cognitive offloading, Productive friction, Sycophancy	Mollick	essay	Nontechnical	free	No login	unknown	2026-10-16
Against "Brain Damage" Garden Mollick pushes back on headlines claiming AI harms our brains and says the danger is thoughtless use. His practical advice is to come up with your own ideas and drafts first, then bring in AI as a reader, editor or devil's advocate, partly because ideas generated with AI tend to look alike. Explains: Cognitive offloading, Productive friction, AI writing tells	Mollick	essay	Nontechnical	free	No login	unknown	2026-10-16
An opinionated guide to which AI to use to do stuff (Summer 2026 Edition) Garden Mollick's mid-2026 buyer's guide for people who aren't developers. He matches tools to jobs: free chatbots for low-stakes questions, the strongest paid models for second opinions, cloud agents such as ChatGPT Work and Claude Cowork for delegated tasks, and local agents such as Codex and Claude Code for large projects. He also says how to start safely: connect only apps you're comfortable with, and keep approvals on for anything that sends, spends or deletes. Explains: Model choice, Agents, Tools and connectors, Human in the loop, Privacy and data	Mollick	guide	Nontechnical	free	No login	unknown	2026-10-16
Assigning AI: Seven Approaches for Students, with Prompts Garden Describes seven roles you can give an AI, such as mentor, tutor, coach, teammate, student, simulator and tool, each with example prompts and the risks to watch for. Written for classrooms, but the roles work for anyone deciding how AI should help. Explains: Prompting, Custom instructions, Human in the loop	Mollick	paper	Mixed	free	No login	CC BY-NC-SA 4.0	2026-10-16
Centaurs and Cyborgs on the Jagged Frontier Garden Mollick's plain-language account of a 2023 experiment with 758 Boston Consulting Group consultants. With GPT-4, they finished more tasks, faster and to a higher standard, but on one task the AI got wrong they did worse than colleagues working alone. The post names two ways of working well with AI: centaurs split tasks between person and AI, and cyborgs weave the two together. Explains: Jagged frontier, Human in the loop, Verification, Centaurs and cyborgs	Mollick	essay	Nontechnical	free	No login	unknown	2026-10-16
Choosing to Stay Human Garden Ethan Mollick argues that the risk lies in using AI on autopilot, and pulls together studies showing that skipping effort means skipping learning. He points readers to the tutoring modes now built into chat tools and asks everyone to keep deciding what to hand over and what to keep for themselves. Explains: Cognitive offloading, Productive friction, Jagged frontier	Mollick	essay	Nontechnical	free	No login	unknown	2026-10-16
Co-Intelligence: Living and Working with AI Garden Mollick's 2024 book for general readers on working alongside AI as a coworker, tutor, coach and creative partner. It is best known for four rules, roughly: always invite AI to the table, be the human in the loop, treat AI like a person but tell it what kind of person it is, and assume this is the worst AI you will ever use. A good single read for someone who wants the big picture before practising. Explains: Jagged frontier, Human in the loop, Custom instructions, Generative AI, Centaurs and cyborgs	Mollick	book	Nontechnical	paid: hardcover $30.00 list; ebook and audiobook available	No login	all rights reserved (commercial book)	2026-10-16
Getting started with AI: Good enough prompting Garden A short starter guide for people who feel they need to learn prompt engineering first. Mollick says to treat the AI like a capable new coworker who forgets everything between chats: give context, say who the work is for, ask for many options and keep talking. His main instruction is to spend about ten hours using it on real tasks in your own field, where you can spot its mistakes. Explains: Prompting, Prompt engineering, Context engineering, Hallucination	Mollick	essay	Nontechnical	free	No login	unknown	2026-10-16
Giving your AI a Job Interview Garden Explains why public benchmark scores tell you little about whether a model suits your work, and suggests testing models the way you'd interview a job candidate: give them realistic tasks from your field, grade the answers blind and repeat on judgment calls. In one test he asked several models to rate a dubious business idea ten times each; each model was consistent, but their scores differed by 3 to 4 points out of 10. Explains: Evals, Model choice, Jagged frontier	Mollick	essay	Mixed	free	No login	unknown	2026-10-16
Instructors as Innovators: A future-focused approach to new AI learning opportunities, with prompts Garden A 76-page guide with ready-to-use prompts for AI exercises: simulations, mentoring, coaching and co-creation, each with notes on how to run it, how to assess it and what can go wrong. It also includes "blueprint" prompts that interview you and then write a custom prompt for your own topic. Written for teachers, but useful to anyone who trains or coaches colleagues. Explains: Prompting, Custom instructions, Custom assistants, Productive friction	Mollick	paper	Mixed	free	No login	CC-BY-4.0	2026-10-16
Making AI Work: Leadership, Lab, and Crowd Garden A plan for bringing AI into an organisation through three groups. Leaders describe what an AI-enabled workplace will look like and promise that sharing productivity gains won't cost people their jobs. Employees experiment in their own work and share what works. A small lab team turns the best of it into tools and tests. He points out that many workers use AI without telling anyone, and the organisation can't learn from use it doesn't know about. Explains: Automation, Evals, Custom assistants, Adoption rituals over mandates	Mollick	essay	Mixed	free	No login	unknown	2026-10-16
Management as AI superpower Garden Argues that working with AI agents mostly uses ordinary management skills: delegating, giving clear instructions and judging the work. Mollick offers a quick test for whether to delegate (how long the task takes you, how often the AI gets it right, how long asking and checking take) and lists what a good handover covers: the goal and why, limits on authority, what done looks like, checkpoints and a self-check. Includes a class where executive MBA students, most of whom had never coded, built working startup prototypes in four days. Explains: Agents, Steering, Human in the loop, Vibe coding, Verification, Allocation economy, Delegating to AI	Mollick	essay	Nontechnical	free	No login	unknown	2026-10-16
On Working with Wizards Garden As AI produces larger pieces of finished work, you can see less of how it got there, and checking every detail can take as long as doing the job. Mollick suggests deciding when to use AI as a partner you talk with versus a "wizard" you hand a whole task to, judging the result, and asking whether it is good enough for this purpose. He also warns that leaning on AI can wear away the expertise you need to check it. Explains: Verification, Human in the loop, Cognitive offloading, Agents, Delegating to AI	Mollick	essay	Nontechnical	free	No login	unknown	2026-10-16
The Cybernetic Teammate Garden Reports a randomised trial with 776 Procter & Gamble professionals working on real product development tasks. One person working with AI did about as well as a two-person team without it, and teams with AI were the most likely to produce top-ranked ideas. AI users also finished 12-16% faster and reported feeling more positive. Useful for team leads deciding how AI should fit into group work. Explains: Human in the loop, Jagged frontier	Mollick	essay	Mixed	free	No login	unknown	2026-10-16
Wharton Generative AI Labs Prompt Library Garden Seventeen tested prompts you paste into a chat tool, each of which sets the AI up as a coach, tutor or role-play partner that asks one question at a time. Examples include a devil's advocate for a team decision, a team after-action review, a team charter, a meeting midpoint facilitator, a problem-reframing prompt and builders that help you make your own tutor or simulation. Most were written for teaching, but about half are tagged for professional use. Explains: Prompting, Custom instructions, Productive friction, Clarifying questions, Custom assistants	Mollick	prompt-library	Mixed	free	No login	CC-BY-4.0	2026-10-16
12 Rules to Vibe Code Without Frustration Garden Peter Yang's tips from more than 50 hours of building apps with AI as someone who doesn't code. The rules start before any code: have the AI write a short readme with requirements and a few milestones, ask for a plan and the simplest options first, then build in small steps, test after each change and roll back when something breaks. A good checklist for a first app build. Explains: Vibe coding, Smallest useful version, Context engineering, Verification, Multimodal AI	Peter Yang	essay	Nontechnical	free	No login	unknown	2026-10-16
20 Tips to Master Claude Code in 35 Minutes (Build a Real App) Garden A product manager's tour of Claude Code while building an app. The free part covers planning before code, asking for the simplest option first, going from spec to to-do list to code, and setting up a project memory file with your preferences. Later tips on debugging and version control are paid. Explains: Vibe coding, Custom instructions, Memory, Agents	Peter Yang	video tutorial	Mixed	partly paid: first six tips free, rest needs a paid subscription (price not checked)	No login	unknown	2026-10-16
Behind the Craft — Transcripts Garden Full transcripts of Peter Yang's YouTube tutorials and interviews (181 episodes when checked), each tagged with title, guest, date and topics, plus a topic index. You can load the folder into an AI tool and ask questions across all of it. Also a tidy example of publishing your own work in a form AI tools can read. Explains: Projects and knowledge, Context engineering, Deep research	Peter Yang	transcript archive	Mixed	free	No login	none stated	2026-10-16
Behind the Craft (YouTube channel) Garden Peter Yang's channel of practical AI tutorials and interviews with people who build AI products. Many tutorials show a full build on screen in under 30 minutes, such as a slide deck, a research agent or a small app, which makes them easy to follow along with. Explains: Vibe coding, Agents, Skills, Automation	Peter Yang	video channel	Nontechnical	free	No login	unknown	2026-10-16
Build Beautiful Slides with Claude Code in 12 Minutes (12 Formats + 3 Templates) Garden Shows a slide-making skill that takes a rough outline, asks you questions about the content, looks up gaps, builds the deck as a single web page and then screenshots each slide to fix its own layout problems. The habit worth taking away is the last step: have the AI look at what it made and fix it before you see it. Explains: Skills, Clarifying questions, Artifacts and canvases, Verification	Peter Yang	video tutorial	Nontechnical	free to read; the /slides skill needs a paid Behind the Craft membership (price not checked)	No login	unknown	2026-10-16
ChatGPT vs Claude vs Gemini: The Best AI Model for Each Use Case in 2025 Garden Peter Yang gives the same tasks to three AI assistants, such as editing his own draft from his writing samples and researching a market, and says which one he would use for each kind of work. The verdicts are from mid-2025 and will age, but the method is worth copying: one real task, the same prompt to each, and a judgement you write down. Explains: Model choice, Deep research, Jagged frontier	Peter Yang	essay	Nontechnical	free	No login	unknown	2026-10-16
Introducing My New AI Learning Path and Co-Pilot Garden A starting map to Peter Yang's tutorials, sorted into six tracks: a personal AI assistant, career, saving time at work, prototypes and apps, workflows and agents, and building AI products. Each entry is marked if it is a video interview or a paid post. Useful for picking a first tutorial by goal rather than by date. Explains: Custom assistants, Vibe coding, Automation, Agents	Peter Yang	reading list	Nontechnical	free post; about half the linked posts and the AI co-pilot need a paid subscription (price not checked)	No login	unknown	2026-10-16
n8n Tutorial: Build an AI Workflow to Extract Top Insights From All Your Newsletter Emails Garden Builds a weekly automation in n8n that collects every email you have labelled as a newsletter, passes them to an AI model with a summary prompt and emails you one digest on Sunday morning. A concrete first automation for anyone drowning in subscriptions, with each node's settings spelled out. Explains: Automation, Scheduled tasks, Tools and connectors, Context window	Peter Yang	tutorial	Nontechnical	paid post: setup steps 1-4 visible free, the prompt and final step need a paid subscription (price not checked)	No login	unknown	2026-10-16
Use My /Human-Review Skill to Edit HTML and Markdown Files Visually Garden Introduces a free skill that opens a document or web page your AI made in the browser, where you edit text directly and leave comments as you would in a shared doc, then send all of it to the agent in one batch. It replaces long chat messages like "in the third paragraph, change X". For anyone reviewing AI-made plans, pages or copy. Explains: Human in the loop, Artifacts and canvases, Skills, Steering	Peter Yang	tool tutorial	Nontechnical	free	No login	unknown	2026-10-16
Generative AI without guardrails can harm learning Garden A field experiment with about a thousand high-school maths students. Plain ChatGPT helped them during practice but left them doing worse on the exam than students without it, because they used it as a crutch. A tutor version with guardrails removed the harm but did not improve results, so guardrails alone don't make people learn. Explains: Cognitive offloading, Productive friction	PNAS	paper	Mixed	unknown	Unknown	unknown	unknown
Making things hard on yourself, but in a good way: Creating desirable difficulties to enhance learning Garden A classic, readable chapter from learning science on why some struggle helps you remember: spacing practice out, mixing topics, testing yourself and producing answers before looking them up. Difficulty only helps if you can still succeed, so friction with AI should match the person. Explains: Productive friction, Cognitive offloading	UCLA Bjork Lab	book chapter	Mixed	free	No login	unknown	unknown

Machine-readable: /catalogue.json. Found something useful that isn't here? Suggest it.