upside down •|• illustrative writer •|• pow in hi-lights •|• social worker @itplaysout
cum find out
Joined January 2024
- Tweets34.9K
- Following997
- Followers1.3K
- Likes43.6K
Pinned Tweet
i use excalidraw :
“rochas has been traumatized”
he said he’s seen A LOT
of standard talk on the tl following the nft cooks.
he’s seen:
➛ walewoosh “cryptic” post about it
➛ deebs quote of that wales "cryptic post"
➛ big chog “claiming” he saw it before wale 😂
➛ central bank here. 1000 charter there.
➛ and a whole lotta bros & queens chattering about standard.
wtf is standard? what’s all za fuss about?
thread 🧵
robin | retweeted
Kippo social media design already taking shape
Anticipate premium design that feeds the eyes
First ever project on Figma
Photshop deportee
robin | retweeted
𝟐𝟒 𝐃𝐚𝐲𝐬 𝐨𝐟 𝐒𝐞𝐜𝐮𝐫𝐢𝐭𝐲 𝐰𝐢𝐭𝐡 𝐂𝐲𝐛𝐞𝐫 𝐋𝐮𝐯 #𝟎𝟎𝟗
𝐏𝐨𝐬𝐭 𝐭𝐡𝐞 𝐏𝐢𝐜𝐭𝐮𝐫𝐞𝐬, 𝐍𝐨𝐭 𝐘𝐨𝐮𝐫 𝐋𝐢𝐯𝐞 𝐋𝐨𝐜𝐚𝐭𝐢𝐨𝐧.
You’re out with friends, visiting a new place or enjoying a solo date.
You take pictures, post them online and even tag your location.
Sharing your experiences online feels harmless, but posting while you’re still at a location reveals where you are in real time.
Repeated updates could also reveal places you visit regularly, your daily routines and periods when you’re away from home.
𝐇𝐨𝐰 𝐭𝐨 𝐬𝐡𝐚𝐫𝐞 𝐦𝐨𝐫𝐞 𝐬𝐚𝐟𝐞𝐥𝐲:
♡ Share pictures after leaving the location, especially when travelling or visiting somewhere alone.
♡ Avoid posting in real-time and sharing details that reveal your exact whereabouts.
♡ Review your camera and social media location settings. Disable location tagging when you don’t need it.
♡ Check your pictures for clues such as street signs, house numbers and identifiable landmarks that reveal your location.
♡ Review who sees your posts and be mindful of people who tag you or share your whereabouts without your permission.
𝐓𝐨𝐝𝐚𝐲’𝐬 𝐭𝐚𝐬𝐤: Review your location and tagging settings on social media. Make a habit of posting location-specific updates after you’ve left.
Your followers don’t need to know where you are while you’re still there.
#CybersecurityAwarenessMonth
Till next time,
Cyber Luv 👩🏽💻💘
𝟐𝟒 𝐃𝐚𝐲𝐬 𝐨𝐟 𝐒𝐞𝐜𝐮𝐫𝐢𝐭𝐲 𝐰𝐢𝐭𝐡 𝐂𝐲𝐛𝐞𝐫 𝐋𝐮𝐯 #𝟎𝟎𝟖
𝐘𝐨𝐮𝐫 𝐖𝐚𝐥𝐥𝐞𝐭 𝐀𝐩𝐩𝐫𝐨𝐯𝐚𝐥 𝐌𝐢𝐠𝐡𝐭 𝐎𝐮𝐭𝐥𝐢𝐯𝐞 𝐭𝐡𝐞 𝐓𝐫𝐚𝐧𝐬𝐚𝐜𝐭𝐢𝐨𝐧.
A few weeks ago, I connected my wallet to a decentralised exchange to swap some tokens.
I approved the token spending request, completed my swap, and moved on with my day, unaware the approval could remain active long after the transaction.
A token approval gives a smart contract permission to spend a specified amount of a particular token from my wallet.
Some approvals have a spending limit, while others grant unlimited spending permission.
An unlimited approval has no set spending cap. If the approved contract is malicious or gets compromised, my tokens could be at risk.
𝐇𝐞𝐫𝐞’𝐬 𝐡𝐨𝐰 𝐭𝐨 𝐩𝐫𝐨𝐭𝐞𝐜𝐭 𝐲𝐨𝐮𝐫𝐬𝐞𝐥𝐟:
♡ Avoid unlimited approvals when a smaller allowance meets your needs.
♡ Review old approvals, including those for dApps you no longer use. Stopping the use of a dApp does not automatically revoke its permissions.
♡ Visit revoke.cash to review your token approvals and revoke permissions you no longer need. Check the website address and selected network before connecting your wallet.
♡ Remember that revoking an approval requires an onchain transaction and usually a gas fee. It prevents future spending through the allowance, but does not recover tokens already stolen.
𝐓𝐨𝐝𝐚𝐲’𝐬 𝐭𝐚𝐬𝐤: Visit revoke(dot)cash and review your wallet approvals. Revoke any permissions you no longer need.
A completed transaction does not mean its spending approval has expired.
#CybersecurityAwarenessMonth
Till next time,
Cyber Luv 👩🏽💻💘
robin | retweeted
for the love of “where did it come from”
reading this jolted me
there’s an abundance of data used to train ai agents today, but data is only as trustworthy as its origin. knowin’ where it comes from helps us verify its integrity & avoid training agents on bad data. - hence why i love the words in the ss
think of it like food. produce comes from farms, but not every farm grows organic food. if we want to know what we’re eating, we need to trace its origins and verify how it was produced. imagine a mall that sells both organic and fertilizer grown produce. trace would be the mall’s (datafdn) verification system, checkin’ every contribution to ensure the right produce reaches the right buyers, while walrus would be the storage house where the produce is kept
this also reminds me of that hulk scene where rad!oactive blood contaminates a drink, gets shipped out, & an unsuspecting old man drinks it. a single failure in that supply chain caused fatal consequences. the same principle applies to ai: bad data can contaminate an entire system, making provenance & integrity essential.
data is life mehn… glad this article highlights the enormous importance of handling, verifying, tracing and organising human data properly.
cos in the end, it’s not just about the data we feed ai, but knowing what we’re feeding it, where it came from, and whether it deserves to be trusted.
salud!
robin | retweeted
One thing about Polymarket that can catch you off guard is that the price you see isn't always the price you're actually getting.
The fee is usually packaged into your trade-slip payout, so it's easy to overlook
that's why i've been using Thumper, a free Chrome extension from @magicmarkets.
You open any Polymarket sports market, and it shows you Magic's price right beside it.
Now you can compare both before placing a bet instead of just going with what you see first
and Magic charges zero commission too.
It's a small tool yeah, but it can make a difference over time.
🤝 Paid partnership
1/ i built a chrome extension for @magicmarkets that shows you when polymarket isn't the cheapest place to take a sports position
it's called thumper. Free to use, no account required 🧵
from my observation,
for the lil gae side, the men shaping the fashion world’s most iconic designs are often men, who particularly lean more into their feminine side. luxury law (zendaya dresser) and a few others come to mind
as for the lil sch!zophrenic bit, i believe alte fashion demands a certain creative madness: the ability to envision realities beyond the ordinary & craft pieces that shatter societal conventions. tapping into ur unconventional side to create sumtin so distinctive, it makes humans pause & go, woww. like me wearing a skirt- like outfit design to class. sumtimes, originality demands breaking the familiar rules entirely. - this also extends outside fashion, too
so yeah, that’s how i see it.
robin | retweeted
if you’re out of ideas
- wash the dishes
- iron ur shirts
- take a long walk
- kick rocks
- fool around
- wander aimlessly
- scrub your toilet
- stare at a spot on the wall for as long as possible
most best ideas often come wen your bored - take time to be bored.
robin | retweeted
daniel finally got his hands on some GEN and started thinking about staking.
in his head, the plan was simple. lock his tokens, earn some rewards, and let them work while he focused on other things.
easy money, right? 😅
well, not exactly.
bcos daniel hadn’t asked himself one important question.
who was he trusting with his stake?
let’s talk about it. 🧵
robin | retweeted
gm frens
I built Chogchemy: turn your @ChogNFT into memes, give it a family comic, and keep its creative history in one passport.
Built for #Chogathon 🎨
Try it: chogchemy.up.railway.app
for the love of “where did it come from”
reading this jolted me
there’s an abundance of data used to train ai agents today, but data is only as trustworthy as its origin. knowin’ where it comes from helps us verify its integrity & avoid training agents on bad data. - hence why i love the words in the ss
think of it like food. produce comes from farms, but not every farm grows organic food. if we want to know what we’re eating, we need to trace its origins and verify how it was produced. imagine a mall that sells both organic and fertilizer grown produce. trace would be the mall’s (datafdn) verification system, checkin’ every contribution to ensure the right produce reaches the right buyers, while walrus would be the storage house where the produce is kept
this also reminds me of that hulk scene where rad!oactive blood contaminates a drink, gets shipped out, & an unsuspecting old man drinks it. a single failure in that supply chain caused fatal consequences. the same principle applies to ai: bad data can contaminate an entire system, making provenance & integrity essential.
data is life mehn… glad this article highlights the enormous importance of handling, verifying, tracing and organising human data properly.
cos in the end, it’s not just about the data we feed ai, but knowing what we’re feeding it, where it came from, and whether it deserves to be trusted.
salud!
robin | retweeted
Don't ever let them doubt you.
@itplaysout now LIVE on @careem // our new super app home in MENA.
Network Ignite. Play. Imminent.
Next...
if you’re out of ideas
- wash the dishes
- iron ur shirts
- take a long walk
- kick rocks
- fool around
- wander aimlessly
- scrub your toilet
- stare at a spot on the wall for as long as possible
most best ideas often come wen your bored - take time to be bored.
robin | retweeted
if you want to never hit your session limits on Claude/Codex,
bookmark and use these 10 tools.
1 → RTK - github.com/rtk-ai/rtk compresses terminal output from commands like git, grep and tests before it reaches your coding agent.
2 → Headroom - github.com/headroomlabs-ai/h…
compresses tool outputs, logs and other context so agents process less irrelevant information.
3 → Graphify - github.com/Graphify-Labs/gra…
turns your codebase into a queryable knowledge graph, helping agents find relevant code without repeatedly reading everything.
4 → Context Mode - github.com/mksglu/context-mo…
keeps bulky tool outputs out of your main context and preserves useful information locally.
5 → Claude Code usage guide - support.claude.com/en/articl…
learn how to manage context with /clear, /compact and /context (I will advise you turn it into a skill)
6 → Aider Repo Map - aider.chat/docs/repomap.html
gives your coding agent a compact structural map of your repository instead of dumping entire files into context.
7 → OpenAI Prompt Caching - developers.openai.com/api/do…
reduces costs on repeated input tokens when building with supported OpenAI APIs.
8 → OpenAI Compaction - developers.openai.com/api/do…
condenses long API conversations while preserving useful context for the next stage.
9 → ccusage - github.com/ryoppippi/ccusage
inspect Claude Code usage and token statistics so you can understand where your usage goes.
10 → LLMLingua - github.com/microsoft/LLMLing…
compresses long prompts by removing less important information before sending them to a model.
robin | retweeted
about 6 months ago i had a crazy experience with a friend.
it was an argument and irl i hate argument’s and confrontations but the argument was about something i was extremely knowledgeable about
this argument in question was about why we're all here every morning posting “gm” and asking “what's the alpha” 24 hours a day on an app with a black background and one letter as it's logo.
yes, you guessed right……it was about crypto
truth is i kinda lost (begins drawing in sand)
but don't crucify me yet, i can explain🧵🔻
