Anthropic's developers share tips, playbooks and opinions on building with Claude and Claude Code, including agents, skills, context engineering and prompt caching.
Why I recommend it: Written by the company that makes Claude, so it's practical but not neutral. Best read by people who already use Claude for coding.
A long explainer on what an "agent harness" is — the loop, tools, memory, sandbox, permissions and budget caps wrapped around an AI model — its six core components, and why the wrapper now matters more than which model you pick. Opens with an agent that burned $400 in API calls overnight in a retry loop.
Why I recommend it: Free to read, no sign-up, and the best single walkthrough of the six pieces if you are trying to understand why agents that demo well fall over in real use. Two honest caveats: it is published by a consultancy that sells help with exactly this, so the framing favors building a harness; and the byline on it is not a person I can verify anywhere independently, which on this site means treat the article as a useful explainer rather than a sourced authority. For neutral definitions, the Wikipedia entry on agent harness and Addy Osmani's write-up cover the same ground.
The first volume of Tech:NYC's Decoded Futures prompt cookbook: ready-to-copy AI prompt recipes for nonprofit work, sorted by task, each offered at three levels of detail — a quick one-paragraph version, a structured version with steps, and a full version specifying role, steps and output.
Why I recommend it: Free, no account, nothing to install — you copy a prompt and paste it into whichever assistant you already use. Start at the Low level if you are new; it works with no setup. Swap the bracketed placeholders for your real organization details, because a prompt left generic gives you a generic answer. Volume 2 is also in this library; V.1 is still worth keeping for the tasks it covers that V.2 dropped. Written for nonprofit teams, but the recipes for writing, summarising and planning transfer to any job.
A free, no-code recipe book of practical AI prompts built for nonprofit teams. Turn existing reports, events, and materials into slide decks, web pages, plain-language translations, social copy, and repeatable workflows. Every recipe includes a privacy badge so you know what stays on your computer, what goes public, and what connects to a vendor account.
Why I recommend it: Free to use. Built by Decoded Futures (a TechNYC program). The recipes are framed for nonprofit work, but the same patterns work for job searches, career content, and small-business tasks.
A three-page printable reference for Google Gemini: the PART prompt formula (Persona, Action, Reference, Tone), follow-up prompting, when to use Flash, Pro and Thinking models, saving your own context in settings, and worked sample prompts for Gemini inside Chrome, Gmail, Docs, Sheets, Slides and Drive. Includes a plain warning about Gemini being confidently wrong.
Why I recommend it: Print this and keep it next to you for a week. Most people get weak AI output because they skip the persona and the reference, and this one page fixes that faster than any course will. The limitations box at the bottom is the part to take seriously.
Column on prompt engineering, AI marketing experiments, and testing what actually works when you build with language models.
Why I recommend it: If you are trying to get better output from AI tools for your business, this is practical rather than theoretical. Steal the experiments.
21 structured lessons on prompt engineering and building generative AI applications, with practical exercises in Python and TypeScript.
Why I recommend it: The best free course for actually building something. Work one lesson at a time and keep the code you write — that is your proof of skill.
Experis/ManpowerGroup research summary on how employers and employees are actually using AI at work, plus a five-step action plan for building AI career durability.
In plain terms: A research write-up from staffing firm Experis showing most employers now use AI in hiring and are fine with candidates using it too, while very few companies have AI fully rolled out. It argues AI mostly augments jobs rather than replacing them, and lists five practical steps — build durable skills, learn your company's AI tools, research use cases for your role, take free training, and propose a small pilot.
From the site: Exploring the key findings of our new report: Building and Sustaining a Meaningful Career in the AI Age.
Why I recommend it: The headline stat matters for job seekers: 85% of employers say it is fine for candidates to use AI during hiring, and 53% already use AI in hiring and onboarding. Use the five-step durability plan as a checklist — learn what AI your employer is deploying, find use cases for your role, take free training, then pitch one small pilot you can measure. That pilot becomes a resume bullet.
