Hayz Labs · https://hayzlabs.com/browse/
81 entries · newest first
Browse everything
Every guide, workflow, tool and setup page in one place. Filter by type, level or theme.
Type
Level
Theme
81 entries
Newest first
Open-source AI tools: one pick for each jobEight everyday AI jobs, from chatting with a model on your computer to testing prompts, each with one open-source pick from our Top 20 lists, one alternative, what it needs and how to start.AI at the doctor's: the listening scribe, AI-drafted replies, and checking your visit noteWhat to ask when your clinic uses an AI scribe or AI-drafted portal replies, a checklist for reading your visit note, and a copy-paste request to correct a wrong note.AI glasses: what do they record, which settings matter, and what can I do if I'm filmed?What camera glasses with a built-in assistant capture and where it goes, an owner's settings check, and what a bystander can do if they are filmed, with the reported privacy problems labelled.An AI answered the phone: how do I reach a person, and should I let an AI call for me?Phrases and steps that get you past an AI receptionist to a person, what to get confirmed in writing, where the law says a bot must say it is a bot, and a short approval list if you let your phone's AI make a call for you.Coding agents for people who don't write codeWhat a coding agent is, five safe first projects you can check without reading code, the settings to start with, and a first prompt that keeps you in control.Connect your medical records to an AI? A decision and a setup checkA yes or no check for connecting your medical records or health apps to an AI assistant such as ChatGPT Health, a six-item setup check if you go ahead, and a routine to disconnect and delete.Interviewed or screened by AI: what to expect, how to prepare, how to ask for a personHow to find out which kind of AI interview you face, a copy-paste email asking what it judges and whether a person can review it, a 10-minute check before you start, and the rules that give some candidates a notice or a human review.Is AI coming for my job? A 30-minute task-by-task check and a skills planBreak your job into its real tasks, mark which ones AI does now, helps with or can't do yet, read what the studies actually measured, and pick two skills to practise for 30 days.Is the app I made with AI safe to share? A 20-minute check before you send the linkSix checks to run before you send people a link to an app you built with Lovable, Replit, Bolt or a similar builder: what it stores, who can read each table, secret keys, and a test as a stranger.Job search and interview prepBuild one experience file you wrote yourself, then for each application map it to the job ad, tailor without inventing anything, check the company, rehearse with a model that grades honestly, and follow up within a day.Kids, homework and a family AI planHouse rules by age, the hints-not-answers setup that avoided the exam drop in one trial, and a one-page family plan to review each term.Letting an agent near your email, browser and filesNever give one agent private data, untrusted content and a way to send things out at once without a person confirming the send. A risk table, a five-rung permission ladder, a plain guide to tools and connectors and an eight-item setup check show how.Research a topic with citation checkingA short brief in which every claim links to a source you opened and checked, built by letting a model research, a script check links and quotes, and a person read what matters.Someone made a fake nude or sexual image of me (or my child): what do I do in the first 48 hours?A calm first-hour checklist, an evidence log, which free removal tool fits an adult or a child, and how the US 48-hour removal rule works, as of October 2026.Switching AI assistants: how do I take my history with me?What moves between ChatGPT, Claude and Gemini as of October 2026, where to export from each one, how to bring across only what is still true, and a cancel-day checklist for closing the old account properly.Talking to AI when you're struggling: what it can help with, and where it stopsWhat a general assistant is good for when you are stressed, worried or lonely, where it stops, a setup prompt that asks for honesty and real people, and signs it is time to step back.Teaching with AI: how do I plan and give feedback without breaking my school's rules?A weekly routine for teachers who use AI to plan lessons and draft feedback, a student-privacy checklist, and the official guidance behind it from the US and England.What not to pasteA three-tier rule and a two-minute settings check, so nothing you regret goes into a chat.Ads and sponsored answers in AI chats: how to spot them and get a straight recommendationWhere ChatGPT, Google and Copilot place ads and how they are labelled, why an unlabelled answer can still be steered, a prompt that asks for criteria, sources and any paid link, and a worked check.Privacy settings pass: 15 minutes, once a quarterA quarterly pass through training, retention, memory and connectors in every assistant you use, with each vendor's own settings as read between 2026-10-04 and 2026-10-10.AI answers or search results: when to use whichA routing table for quick facts, recent facts and research, a 60-second citation check, and a prompt for assistants that search the web, with three 2025 studies on fewer clicks and wrong citations.AI detectors: what they can and can't tell youA detector flag is a guess about a text and no proof about a person; independent tests found false positives and unreliable results, and a 2025 paper found some tools much better, so a flag is a reason to ask.AI memory: what it keeps, and a ten-minute auditWhat ChatGPT, Claude and Gemini say they remember across chats, how to see it, and a ten-minute routine to list it, prune it and choose a memory setting for each kind of task.AI voice scams: hang up and call back on a number you knowA call-back rule on a number you already have, a short card for older relatives, and an optional family phrase as an extra check, with the official 2025 advice behind it.AI-written phishing: stop looking for typos, set up passkeysScam emails no longer give themselves away with bad spelling. A 20-minute setup that protects your accounts even when you cannot tell a fake from the real thing.Am I actually faster?Median minutes with and without AI, including review and fixes, from your own runs, with a quality check beside the time.Asking questions of your data when you're not an analystA 34-minute routine for asking an assistant about a spreadsheet: check the row count and a known total first, test the question on a sample you worked out by hand, then reconcile the answer and keep the code.Build a small agent yourself: one tool, a step limit and a logAn agent is a model calling tools in a loop. Build the smallest useful one: a read-only search over a folder of notes, a hard limit of six steps, a token and cost tally, a log of every tool call, and answers that must cite their files. Then test it on nine questions, one aimed at a planted instruction.Build an assistant other people will useGive a shared Claude Project, Gemini Gem or skill, or a custom GPT in a work workspace one job, assume anyone who can chat with it can read its instructions and files, and pass a written 15 to 20 question test before anyone relies on it.Check a claim you saw online, with AI as a helper, not the judgeWrite down exactly what a post, screenshot or headline claims, ask an assistant to split it into checkable parts instead of ruling on it, then leave the page, find the original and look for an existing fact check.Check before you trust: a five-minute routineA seven-line checklist you run on any AI answer before you act on it.Check the request, not the voice: payment changes at workA short check for any request to change bank details or make an unusual payment, so a convincing fake call, message or video call cannot move money on its own.Checking agent work: review loops and work you didn't watchAn eight-step routine for checking what an agent did while you were not watching: read the log first, test its claims against the files, sample by stakes, use a different checker, and know when to take the work back. Plus a one-off test with five attack inputs before you rely on a workflow you will run again.College and university applications: what AI use is allowed, and how to use it honestlyWhat the Common App, UCAS and four universities say about AI in application essays, which help is usually fine, a feedback-only prompt, and keeping evidence of your process.Complain about a faulty product or poor service, with AI drafting and you checking your rightsGather the facts, quote your rights from an official page, have an assistant draft a short, firm letter, check every legal claim, and know the next step if the company says no.Describe the failure: reports an agent can act onTurn 'it's wrong' into a seven-line report that an assistant or an agent can fix in one pass, and a rule for when to stop correcting and start fresh.Do I need to pay for AI, and for which plan?When a free plan is enough, when one paid plan is worth about $20 a month, how to pick that one plan, and a two to four week test before and after you pay.Find near-duplicates and group a long list by meaning (embeddings)Turn each item in a long list into numbers with an embeddings model, list the closest pairs as near-duplicate candidates, group the rest with a short numpy script, and read samples yourself before you name or count anything.Fix a computer, phone or app problem with AI, safelyDescribe the problem precisely, ask for the safest steps one at a time with an undo for each, back up before anything risky, never run a command you can't explain, and know when to stop and call the maker.Get a spreadsheet formula from AI, then prove it worksDescribe the sheet exactly, ask for the formula with its edge cases, run it on 8 to 12 rows you worked out by hand, check for silent failures, and only then fill down and reconcile one known total.Get data you can trust from an API: schemas, validation and retriesValid JSON, JSON that matches your schema and JSON with the right values are three different things. Let the vendor guarantee the shape, check the values in your own code, retry once with the errors, then hand the item to a person.Getting disagreement instead of agreementA standing instruction and five prompt cards that get you criticism instead of applause when you ask an assistant to check a plan, a draft or a decision.Health, money and legal questions: green lights and red linesWhat an assistant can safely do with a health, money or legal question, where you stop, who to ask instead, and how to ask for a person to review a decision software may have made about you.Help an older relative start using AI, safelyOne sitting together: install the real app, set up the account with 2-step verification, try three tasks that matter to them, and leave a card of rules by the phone. Then check in after a week.Learn to code with AI without skipping the learningChatGPT or Claude matters less than how you set it up. Make it a tutor, write each piece yourself first, use it to explain errors and code, and test yourself without it every week.Learning with AI without losing the learningThe same kind of tool more than doubled learning gains in one trial and cut exam scores in another. Four setup rules of ours and a checklist, built from those trials, aim for the first.Local or cloud: when a model on your own device is enoughDecide by what the data is and what the task needs, and check by hand whether a model will fit in your computer's memory, using a rule of thumb of about 0.6 GB per billion parameters.Long agent jobs: pieces, milestones and stop rulesA routine for cutting a big job into checkable pieces for an agent, with a milestone file for each piece, a spend and time cap, a log, and a stop rule that tells you when to take the work back.Make a presentation with AI: your outline first, then check every slideWrite the point and the audience yourself, get an outline of sentence headlines from your own notes, edit it, then let the slide tool build the deck and check every fact, image and slide before you present.Make an image with AI: describe it well, check it, and know what you can use it forA prompt template for posters, slide pictures, illustrations and mockups, one change per edit, a check for text, hands, logos, invented facts and real people, and what the labels, vendor terms and US copyright guidance say.Meeting notes to decisions and ownersA message of decisions, owners and dates from a meeting, each backed by a quote that attendees can confirm.Personal admin with AI: inbox, letters and bills, with you approving every send and paymentA weekly pass where an assistant sorts your mail, explains official letters and checks bills, while you send every reply and make every payment yourself.Photos in, data out: forms, receipts, whiteboardsA 19-minute routine for turning a photo of a receipt, form or whiteboard into a table: transcribe first, extract second, then check the sums outside the model and read every number that matters against the photo.Plan a trip, then verify every stopAn 82-minute routine for a trip: write the brief yourself, ask for two contrasting itineraries in a table with an official-site column, check every stop on its own site, use government advice for entry rules, and book it yourself.Plan a week of meals and a shopping list with AI, and check what mattersA household brief, a 5 to 7 day dinner plan that reuses ingredients, and a shopping list by aisle, then the checks that matter: allergen labels, official cooking temperatures and real prices.Point your phone at it: what AI can and can't tell you from a photoA traffic light for photo questions: when to trust what your phone's AI says about a plant, a pill, a rash, an error screen or a menu, and what to do instead.Practise a new language with AI: speak daily, get corrected, track progressSet a level and a goal, give the assistant one standing instruction, talk for 15 minutes a day, review your repeated errors weekly, and check with a person once a month.Prepare for a negotiation with AI: a pay rise, a job offer or a priceGet the numbers yourself from dated, named sources, write your target, walk-away point and alternative, then use an assistant as a tough practice partner, an objection finder, a follow-up drafter and a debrief coach.Prompt, retrieve or fine-tune? Getting a model to use your materialName the problem first. Missing knowledge is fixed by putting your material in front of the model. Wrong format or style is fixed by examples, and only rarely by fine-tuning. Climb one rung at a time and measure each rung on the same 20 questions.Prompting in one pageThe few prompt habits that vendor docs and published tests agree on, the tricks that make no reliable difference, and how to save a prompt you reuse and change it without breaking it.Read a contract, lease or terms with AI before you signUse an assistant to understand a lease, job contract or phone plan before you sign: remove personal details, ask for a summary with every key clause quoted and numbered, check each quote against the document, then take your questions to the other party or an adviser.Read the source, use the summary as a mapA 45-minute routine for a paper or report that matters: ask the assistant for a map, read the sections that count in the original, write what you remember, then check your notes against the text.Reconcile two exportsTwo files that should agree, such as a bank export and your bookkeeping: a small script matches them by rules you approve and proves nothing was lost, you check a sample by hand, and an assistant only helps label the leftovers.Run one prompt over hundreds of rows, then measure how often it's wrongLabel, sort or extract hundreds of rows with one fixed prompt: hand-check a test set first, run in numbered chunks, check the counts mechanically, then hand-check a random sample and turn the errors you find into an error rate with a range.Script, workflow or agent: how much to automateEight rungs from doing it by hand to an unattended agent, five questions to pick the lowest rung that does the job, and the tasks where a plain script beats a model.Start hereFive short steps that leave you with tasks AI already helps with, a privacy rule, a file that tells it who you are and a check routine. Then pick the page for the job in front of you.Talk, then edit: dictation and voice modeSpeak the background and goals, type the exact names and numbers, then check the transcript: what dictation does well, where it invents or drops words, and a prompt that turns a ramble into a brief.Taxes with AI: what it can help with, and what you check on the official siteUse an assistant to explain tax terms and letters, list your documents and sort receipts. Keep identity numbers and full returns out of the chat, and check every rule, figure and deadline on irs.gov or GOV.UK for the right tax year.Tell it who you are, onceA one-page text that stops you re-explaining yourself in every chat.Test any new model on your own tasks (a personal eval set)Keep ten to twelve of your own tasks with written pass rules, run them on any new model or plan, judge them yourself, and switch only by a rule you wrote in advance.The AI already in your phone, email and office apps: what it reads, and how to turn it up or downWhat the AI built into iPhone, Mac, Gmail, Android, Word and Outlook, Samsung phones, WhatsApp and LinkedIn reads, where it runs, and the setting that turns each one down, as the vendors describe it in October 2026.Transcribe a recording, then check it before you quote itGet a fast AI transcript of an interview, lecture, memo or call, then listen to the parts that matter before you quote, search or summarise it.Translating with AI: when it's good enough, and how to checkMatch the check to the stakes, prepare the text, and check a translation you cannot read: back-translation, numbers and negations line by line, a second tool, and a speaker for the lines that matter.Was this image made or changed by AI? A five-minute checkSix steps to run on a photo, screenshot or video before you share it or act on it: provenance, watermarks, reverse search and the claim itself, and why your eye and AI detectors are not enough.When not to use AIA six-question card to read before you open the chat box. Any no means doing the task yourself first, then using the assistant as a reviewer, or skipping it.Where could AI help this week?Score up to 15 recurring tasks by minutes a week, how repeatable each is and how easily you can check the result. The three highest scores are the tasks to try with AI first.Write a speech, toast or eulogy with AI that still sounds like youGather your own stories, ask for structure rather than a finished speech, have the assistant flag clichés and invented details, then check every name and read it aloud against the clock.Writing in a second language, or with dyslexiaUse AI as an editor of your own draft, keep your meaning and your voice, and know what to do if a detector flags you.Writing: brief, draft, checkA 40-minute routine for a piece you must stand behind: write the brief yourself, let a model draft from it, edit for meaning, check the draft against the brief in a fresh chat, and match every number and name before it goes out.Your first API call, with a spending capSet a hard monthly cap on a separate project before you create a key, keep the key in an environment variable, make one call with curl and Python, and turn the usage fields into a cost. Anthropic and OpenAI side by side, Gemini in one line.Your first useful hour: three everyday tasksThree results you can use today: a tighter email, a dated to-do list, and the case against your own choice.
Nothing matches those filters.