Google just gave everyone the power to build their own AI assistant.
No coding. No subscription. 100% free.
Here’s how to set yours up in under 5 minutes 👇
1. Go to Gemini
➡Visit
➡Click "Gem Manager"
➡Hit "New Gem"
Gems are custom AI assistants you design—Google even has pre-made ones like Brainstormer, Career Guide, and Coding Partner. gemini.google.com
2. Name + Brain
➡Give your Gem a name.
➡Write its instructions (what it should do, how it should act).
➡Use the Magic Pencil ✏️ to auto-rewrite and improve them.
You’re now creating your own custom AI assistant. 🛠
3. Feed it knowledge
➡In the “Knowledge” section, upload up to 10 PDFs or images your AI can reference.
➡Remove or replace them anytime.
➡This is how you make your AI smarter than ChatGPT.
4. Test before launch
➡Use the Preview window to see exactly how your AI will respond.
➡Adjust instructions until it feels perfect.
5. Edit anytime
➡Go back to Gem Manager, click Edit, tweak instructions or knowledge.
➡Hit Update and your AI instantly levels up.
APPLE IS QUIETLY DELETING YOUR PHOTOS AND CALLING IT A FEATURE.
"Optimize iPhone Storage" replaces full-res photos with thumbnails.
On a flight with no Wi-Fi? You're sharing a blurry mess.
Turn off iCloud wrong? Originals wiped.
Fix it in 5 settings. Save this now.
1. The Setting That Started Everything
Go to Settings > Photos.
If "Optimize iPhone Storage" is checked, your originals are not on your phone.
They're floating in iCloud as full files.
What's on your device is a compressed preview.
Tap "Download and Keep Originals" instead.
Your real photos start coming home.
2. The Trap When You Cancel iCloud
Most people don't know this.
If you turn off iCloud Photos while "Optimize Storage" is on, your iPhone gives you 30 days to download the originals back.
Miss that window and they're gone from your phone forever.
Before you ever toggle iCloud Photos off, switch to "Download and Keep Originals" first.
Wait until everything finishes syncing.
Then make decisions.
This search engine completely changed how I use AI search.
It pulls from 80% of the internet that Google literally cannot see - legal cases, financial filings, code repos, threat intel, academic papers.
One API key. Zero blind spots.
It's called AnySearch and here's why every AI agent will run on it: 👇
Here's the dirty secret nobody talks about:
Your AI agent is searching the same ~20% of the web that Google indexes.
Business registrations? Blocked.
PubMed research? Locked.
Finnhub financial data? Paywalled.
VirusTotal threat feeds? Separate API.
That's 80% of high-value data your agent never sees.
AnySearch fixes that with one key.
The magic isn't aggregation. It's the routing.
>> Intent Classifier reads your query
>> Matches it across 4 dimensions: domain x subcategory x content type x region
>> Pulls from the most relevant specialized sources
>> Fuses results using Reciprocal Rank Fusion (RRF, K=60)
>> Returns clean Markdown
You don't tell it where to search. It already knows.
I'll give you the 12 Claude prompts for free that can help you earn $500 a day:
The setup is stupid simple.
A laptop. Wi-Fi. A few hours you'd waste on Reels anyway.
That's it. No course. No coach. No "mastermind."
Paste these prompts into Claude. Replace the brackets with your actual skill and audience.
Prompt 1. Find the money in your skill.
"Act as a digital product strategist. My skill is [skill], experience level [beginner/intermediate/expert], audience is [audience]. Give me 15 passive income product ideas. For each: what it is, who buys it, problem solved, build difficulty, time to launch, best platform. Rank from 'easiest to monetize' to 'highest long-term upside.'"
Accidentally overheard an AI researcher from Anthropic at a conference bar.
He didn't mention prompt engineering or context windows once.
Only talked about 3 structural shifts that separate people who get real value from LLMs vs people who don't.
I turned them into Claude prompts. Here they are:
PROMPT 1 : MAP HOW YOU ACTUALLY USE AI
Act as a workflow auditor, not an AI coach.
Ask me 5 questions to map how I currently use LLMs: what tasks I delegate, what I still do manually, where I copy-paste between tools, what I've tried and abandoned.
Don't give advice yet. Just build the map.
One question at a time.
PROMPT 2 : FIND WHAT'S REAL VS WHAT'S THEATER
Here's my AI usage map: [paste]
Now identify:
- What's saving me REAL time (measurable hours back)
- What FEELS productive but isn't (theater)
- What I'm doing manually that an LLM should be doing
No points. No miles. No VPN. No travel hacker bullshit.
Just 10 prompts.
Here's exactly what I typed: ↓
1. The Booking Window Finder
"Analyze historical fare data for [route] over the last 12 months. Tell me the exact day of the week and number of days before departure when prices hit their lowest. Show the data."
Most people book on a Sunday for a Friday flight. That's the most expensive combo possible.
2. The Airport Swap
"List every airport within a 3-hour drive or train ride of [city]. Compare flight prices from each one to [destination] on my dates. Include ground transport costs in the total."
Flying out of the secondary airport saved me $180 once. Train ticket was $22.