Content Alchemist πβ’ Marketer β’ KOL β’ cREator: @re β’ Building on @scribble_dao
DeFi
Joined June 2024
- Tweets47.4K
- Following1.1K
- Followers2.3K
- Likes24.2K
the annoying part of email as a creator has never been the idea. itβs building something that doesnβt break in outlook, stays on brand, and then actually knowing which parts people clicked.
@Migma_AI is basically an AI email marketer that takes a goal or prompt and hands you a full, editable, client-safe email or series. brand kit, visual editor, preflight checks, the usual pain handled.
the new stuff makes it more usable though. click heatmaps so you can see what actually got clicked, plain-language audience segments, remixable emails so youβre not starting from blank, plus countdowns and app integrations.
still need your own voice and a real list. but it cuts out most of the design and compatibility busywork that used to eat the day. worth testing if you send more than occasional newsletters.
migma.ai
Whatβs new in Migma β¨
Click heatmaps. Smarter targeting. A faster engine. Campaigns from your apps.
Plus remixable emails, countdowns, and more integrations.
Explore whatβs new: migma.ai
chatgpt keeps logs of deleted memories for up to 30 days, per openai's own memory faq.
and claude's memory stays inside claude.
so what your ai knows about you belongs to whichever app you typed it into.
there's another way to build it, here's the case. π»
so where i land is simple.
if you want memory in your own bucket or on a public network, that survives switching tools, this is the design to look at.
if you need encryption by default, use memwal or encrypt first.
your memory shouldn't reset every time you change apps.
that's what @LighthouseWeb3 is building with lighthouse memory.
the sdk and mcp server are live: memory.lighthouse.storage
π½πΈβ€πΌ retweeted
This is my demo video for the @PayBox hackathon.
I built PayRun, a PayBox plugin that lets agents pay up to 20 people USDC on Solana in one transaction, with one approval.
In the video, Ada, Tunde and Chi split dinner, drinks and a taxi. PayRun works out the fewest transfers to settle everyone.
Then it builds one batch payment to 3 people, ready for a single approval.
Payroll, team payouts, contest prizes, trips.
Agents can pay one person. It's time they pay everyone at once.
@moonpay
π½πΈβ€πΌ retweeted
Agents can pay for things now. Good. But none of them know how to bargain, and anybody whoβs bought something in a Nigerian market knows the first price is never the price.
So for the PayBox hackathon I built Haggle.
Itβs a plugin. Your agent talks to a seller that speaks Yoruba, Igbo, Hausa and Pidgin. The seller opens at a list price, has a secret minimum, and counters like a real trader. Lowball it and it holds its price. Push too hard and it walks away.
When they agree, the plugin hands you the payment in PayBox for the exact agreed price, in USDC on Base. You approve it or you donβt. Nothing moves before that, and the seller only takes the money after the work is done.
In the video my agent opened at 3 cents against a 5 cent list price and they settled at 4. Small number, but itβs two programs actually negotiating, live.
A few things I was careful about. The sellerβs minimum never leaves the server, not even inside the deal token. Deal tokens are signed and tied to the exact job, so nobody can edit the price. If you offer more than the ask, you still pay the ask.
It has four tools: menu, start, offer, checkout. Only checkout touches money.
Try it yourself, link below. Set a budget and watch your agent haggle.
haggle-seller.ferz-erz5.workβ¦
Built for the @MoonPay PayBox hackathon. #PayBox
This is my demo video for the @PayBox hackathon.
I built PayRun, a PayBox plugin that lets agents pay up to 20 people USDC on Solana in one transaction, with one approval.
In the video, Ada, Tunde and Chi split dinner, drinks and a taxi. PayRun works out the fewest transfers to settle everyone.
Then it builds one batch payment to 3 people, ready for a single approval.
Payroll, team payouts, contest prizes, trips.
Agents can pay one person. It's time they pay everyone at once.
@moonpay
GOOOOD WAY TO END THE WEEEK!
MORE TO COME.
π¨ CITED CREATOR ALERT π¨
@faze_aze took their @MOI_Tech content to Substack and Paragraph π
Their take on participant-centric authority is now being cited by Gemini.
More places to publish means more chances to become the source frens.
Create. Get cited. Get paid. πΈ
ALT Green text highlights "CITED CREATORS" alongside logos for MOI and Scribble, featuring user @faze_aze and platforms Paragraph and Substack.
π½πΈβ€πΌ retweeted
π¨ CITED CREATOR ALERT π¨
@faze_aze took their @MOI_Tech content to Substack and Paragraph π
Their take on participant-centric authority is now being cited by Gemini.
More places to publish means more chances to become the source frens.
Create. Get cited. Get paid. πΈ
ALT Green text highlights "CITED CREATORS" alongside logos for MOI and Scribble, featuring user @faze_aze and platforms Paragraph and Substack.
π½πΈβ€πΌ retweeted
Everybody has been in the argument where one person says they already paid and the other says nothing came in, and it usually ends with a screenshot that either of them could have edited.
I built a small app to take that argument out of the room. You create an invoice, the other person pays it, and the invoice flips to Cleared on a public record anyone can look up.
If the amount comes in wrong, it shows as Flagged, so nobody has to wonder whether the number was off by a little.
It's live on BOT Chain mainnet and works end to end today. Paying is two steps for now, sending the money and then recording it, and getting that down to one is what I'm working on next.
Try it (pick BOT CHAIN at the top): flare-terminal.vercel.app
Contract: 0xF44df427133003aD13a5cBf4Cdcd871554DC8Af2
every token launch happens on a public blockchain, where anyone can read every transaction.
yet most people still go into a launch without knowing who gets what, or on which terms.
@PraxenProtocol wants to fix the part the chain doesn't show, here's the case. π»
the token is early too.
$PRAX is listed as a proposed ticker, and its utility, from feature access to ecosystem incentives, is not finalized.
so the thing to watch is the product, not a chart.
where i land is simple.
transparency can't make every launch fair.
but rules you can read before you buy, and allocations you can check after, would remove a lot of blind trust.
that's what @PraxenProtocol says it's building. now it has to ship.
