Content Strategist & Global Social Media Manager. #demandgen #SaaS #cybersecurity #datastorage #copywriter #socialmedia
Silicon Valley, California
Joined August 2008
- Tweets43K
- Following46.5K
- Followers43.7K
- Likes68K
marymac retweeted
My Grok Bot got me 500,000 views on 𝕏 in 7 days.
Here's the full setup:
0:00 Introducing Reply Window
0:30 What Grok is NOT doing
0:51 Why the first hours matter
1:07 Why I use this daily
1:32 How the X algorithm works
2:00 Job 1 - new post / QT ping
3:30 Job 2 - unreplied count (4h)
4:10 Re-explaining my need for it
Week 1 of Grok @Bot:
Took 5 minutes to setup.
Goal: Get monetized on 𝕏 in November and make more original videos.
> 1 month for $15 ✅
> Setup content database ✅
> Connected my @superx_so account ✅
> Connected X, Notion, Gmail, and Calendar ✅
> Setup multiple X-growth bots (watchers) ✅
> Built a paper-trader to test AI-investing ✅
Total usage: 2%
Video coming tomorrow on how I maximized my reach on X.
What are you cooking with your bots??
ALT Marshall with MacBook showing Chief chat and Content Ideas
marymac retweeted
Here's the full prompt for the Reply Window bot.
This is how I never miss a reply.
Reply with @bot to instantly build it 👇
My Grok Bot got me 500,000 views on 𝕏 in 7 days.
Here's the full setup:
0:00 Introducing Reply Window
0:30 What Grok is NOT doing
0:51 Why the first hours matter
1:07 Why I use this daily
1:32 How the X algorithm works
2:00 Job 1 - new post / QT ping
3:30 Job 2 - unreplied count (4h)
4:10 Re-explaining my need for it
marymac retweeted
So after all the theater, an American company didn’t acquire TikTok?
youtu.be/V95dUteNskY?is=8JTj…
marymac retweeted
HOLY F*CK... I found 21 GitHub projects that can turn Grok Bot into an entire AI company.
We're talking autonomous teams, persistent memory, browser automation, sales departments, trading desks and much more.
Here are 21 projects worth bookmarking:
1. AGENT ORCHESTRATION
• Orca
Run multiple AI coding agents in parallel using isolated workspaces and coordinated workflows.
github.com/stablyai/orca
• OpenMausBot
An open-source AI coworker with agents, connected apps, and its own computer.
github.com/milind-soni/OpenM…
• Open Duet
Build self-hosted AI coworkers with roles, memory, and routines.
github.com/Anil-matcha/open-…
• Rakazo
Create persistent AI teammates and choose the models they use.
github.com/elie222/rakazo
• OpenBot
A framework for AI coworkers with access to browsers, files, and tools.
github.com/CopilotKit/OpenBo…
2. BROWSER & AUTOMATION
• Web Access
Give your AI agents access to the web through browser automation and reusable site workflows.
github.com/eze-is/web-access
• DevSpace
A development workspace project for building and managing AI-powered workflows.
github.com/Waishnav/devspace
• Unbrowse
Turn browser workflows into reusable API calls instead of repeating manual browsing.
github.com/unbrowse-ai/unbro…
• Grok Bot CLI
Manage Grok Bot agents and groups directly from the command line.
github.com/ScriptedAlchemy/g…
3. MEMORY & KNOWLEDGE
• Obsidian Second Brain
Give AI agents persistent memory through an Obsidian knowledge base.
github.com/eugeniughelbur/ob…
• OpenMeld
A project for connecting AI agents and workflows.
github.com/amplifthq/openmel…
• Rome
An open-source agent operating system designed for persistent, compounding workflows.
github.com/rome-os/rome
• GrokBot Field Notes
A collection of agent roles, playbooks, rules, and lessons learned from building Grok Bot teams.
github.com/unicodef1wn/grokb…
4. BUSINESS & TRADING
• OpenCompany
Build an AI-powered company with specialized AI employees.
github.com/zeenie-ai/OpenCom…
• HyperGrok Trading Desk
A trading desk powered by specialist agents for research, risk, and execution.
github.com/galleonlabs/hyper…
• Gojiberry AI Sales OS
An AI outbound sales team with specialized roles and MCP integrations.
github.com/romangojiberryAI/…
• Crew Contract
A framework for defining roles and contracts between agents in a team.
github.com/lsj210001/crew-co…
5. TEMPLATES & SKILLS
• Awesome Grok Bot
A directory of public Grok Bot configurations you can explore and learn from.
github.com/kydlikebtc/awesom…
• Root Agent Skill Framework
A framework for organizing and managing reusable agent skills.
github.com/MrBekoX/root-agen…
• Grok Bot Profiles
A collection of Grok Bot profiles and configurations.
github.com/HAEGONG/grok-bot-…
• Grok Research
Tools and workflows for research with Grok Bot.
github.com/jaredtrichard/gro…
21 projects. 5 categories. A ridiculous amount of things to build.
Bookmark this before you lose it.
Which one are you testing first?
marymac retweeted
i started posting on 𝕏 2 weeks ago. after 3 calls with friends who have 20k-50k followers, my first article hit 150k views.
going viral is easier than it looks.
here's what i learned:
1/ research before you start posting. study creators who regularly get 10k-50k+ views with a much smaller audience than yours. look at their topics, hooks, formats, and what people actually respond to.
2/ split your content roughly 70/30. 70% useful, bookmarkable content. 30% personal thoughts about your niche and quotes on other posts.
to find topics for that 70%, ask yourself:
- what does my audience save? lists, tips, tool stacks?
- what could make their work easier or better?
- what could help them make money?
for personal posts, share your experiences, observations, and opinions. give people a reason to follow YOU beyond the resources.
3/ reply to your audience. help people, have conversations, and listen to their feedback. their questions will tell you what to write next.
4.1/ write 2-4 useful articles around the same questions from point 2.
go deeper than you would in a tweet. include sources, examples, your own research, or workflows you've actually used.
the more useful material you give people, the more reasons they have to quote it or recommend it.
4.2/ after publishing an article, write 3-6 quote posts over the following days.
each one should add something.
a related finding, a useful list, a personal story, or a workflow that builds on the article. give people a different reason to open it each time.
an article that gets 20k views on day one could still reach 300k over the week.
publishing it is only the start.
marymac retweeted
How #algorithm on #X decides which posts to display -- and which posts not to display. From @ryangrowthh.
#demandgen #socialmedia #contentstrategy @VBierschwale @ChiefEngineerCE @PlumbNick @marc_palasciano @LoneStarLegendX @CTechworkers @StevenEdginton @RedPillRabbit @ThomBrady5
marymac retweeted
Cool -->> See popular use cases for #Grok #bot.
#grokbot #ai #agent #usecase #SaaS #developer #webdev #marketer #entrepreneur #founder #sdlc #socialmedia #marketing #contentstrategy #content
Pulled the most popular Grok @bot use-cases over the last 30 days:
ALT Numbered list of 34 popular Grok Bot use cases on a white background
Cool -->> See popular use cases for #Grok #bot.
#grokbot #ai #agent #usecase #SaaS #developer #webdev #marketer #entrepreneur #founder #sdlc #socialmedia #marketing #contentstrategy #content
Pulled the most popular Grok @bot use-cases over the last 30 days:
ALT Numbered list of 34 popular Grok Bot use cases on a white background
marymac retweeted
Explaining how the 𝕏 algorithm (actually) works.
Once you understand this, you can reverse-engineer virality.
0:00 Intro & why listen to me
1:25 How to think about "gamifying" the 𝕏 algo
3:01 The three algorithms: Thunder, Phoenix & Simcluster (intro to algo 1, Thunder)
3:30 Phoenix, Algo 2: the key to going viral
4:20 Simcluster, Algo 3: how niches on 𝕏 work
5:28 The Judge: how 𝕏 scores posts (point system)
8:48 The Bouncer: what gets filtered
11:14 Actionable advice
marymac retweeted
Ambassador applications for @spacexsi / @cursor_ai are open
Details + apply here: anysphere.typeform.com/space…
marymac retweeted
New X algo drops -- see what's new. "Likes still help, and a reply or quote helps even more." #bot #botwars #socialmedia #censorship #x
New X algorithm code dropped last night. Nothing changed in how your posts are scored on X's end.
Here it is in plain English.
1. Mass reporting no longer blurs a post by itself. X deleted the rules that put a blur or warning screen on a post just because it was reported as adult or graphic by a lot of people. Say a post about a crime or a war gets a pile of reports from people who simply don't like it. Before, that alone could hide it. Now it can't. Posts flagged by X's own high-precision checks for gore or adult content still get blurred, and reaction posts only pick up a gore blur from those same checks.
2. The Brazil 2026 election filter is gone. X had a special filter for election accounts there, and with the election over, they removed it. It never touched your posts.
3. The new-account filter is longer. X has a soft filter that hides mildly adult posts from brand-new users. It used to cover a viewer's first day. Now it covers their first 7 days, and only while they've done very little on X. It only affects brand-new users.
4. NOT LIVE YET: two experiments are built but switched off. One builds the feed one slot at a time, like filling seats in a row, using the same weights. The other mixes in some random variety so the feed isn't always the same kind of post. There's also a new filter to pull risky videos out of the video carousel, but the carousel itself is still off.
5. X also confirmed in its notes that ranking is now better at predicting how long people spend on a post, and that the feed is trained to surface posts people read all the way to the end. So write things worth finishing.
The rest is behind the scenes, like training work and a Grok tool that can spot content hidden behind an overlay.
Likes still help, and a reply or quote helps even more.
Why post your content on #X? Because X gets LONGER VISITS than #Insta #Reddit #Facebook #TikTok.
#demandgen #strategy #socialmedia #marketing #entrepreneur #founder #SaaS #contentstrategy #howto
How #algorithm on #X decides which posts to display -- and which posts not to display. From @ryangrowthh.
#demandgen #socialmedia #contentstrategy @VBierschwale @ChiefEngineerCE @PlumbNick @marc_palasciano @LoneStarLegendX @CTechworkers @StevenEdginton @RedPillRabbit @ThomBrady5
marymac retweeted
BREAKING: NEW FY2026 H-1B DATA IS LIVE
THE $100,000 H-1B FEE DID NOTHING TO SLOW NEW APPROVALS.
This covers the fiscal year from October 2025 to September 2026.
The fee started in September 2025. FY2026 was the first full year with it in place.
First-time approvals still rose 10.6%, to 126,985 from 114,804.
Continuing approvals fell 8.0% to 268,191, down from 291,543.
Total approvals came in at 395,176, down just 2.7% from 406,347 in FY2025.
All figures are APPROVED petitions, meaning either a new employee or an employee keeping their job or starting a new one.
That's 395,176 approvals for foreigners on a temporary visa to start or continue to work in America.
marymac retweeted
🦔Hackers hit supermarket freezers again overnight, this time across Belgium, Sweden and a handful of other European countries. They remotely switched units into defrost mode and pushed them as high as 50°C, ruining frozen stock at chains like Delhaize and Coop. It’s the same kind of attack that hit about 700 stores in France three weeks ago. Many of the targeted controllers are old, internet-exposed models that no longer get security updates.
My Take
Hackers have moved past stolen passwords and into the physical world. Last year one attack shut Jaguar Land Rover’s factories for about five weeks, and now grocery freezers are fair game. Every system that went online to save on staff added another way in, and companies keep adding more, all the way down to AI agents that log into accounts for you. AI also makes the attacker’s job faster and cheaper, so I don’t see this slowing down.
Security turns into a cost every business pays one way or another, either upfront or after the disaster. I think the companies that kept a manual backup (someone who can walk over and check the freezer) look a lot smarter over the next few years.
Hedgie
marymac retweeted
Imagine billions of children being taught by cheap "democratized" AIs like this one.
It's bad enough that rich kids will offload their minds to more precise machines.
Even worse, we'll see loads of working class and poor kids get borgtarded "cuz the Future™."
#CVE #Breachprevention Coding models were trained on HUMAN, not #AI, code:
* Update first: Old infrastructure, or insecure at perimeter
* New AI tools, esp. coding, need a security framework. Developers will do more with code, new skills, and more automation.
#threatdetection
Attacks will accelerate from here on - We are there: Six months ago I said Mythos like capability will be widely available. It's happening. A critical vulnerability, a social engineering attack that looks innocuous, or daisy chaining publicly available vulnerabilities and conducting a sniper strike is all being witnessed in the wild. Incident responders have never been busier.
Some key points:
- attacks are not becoming more sophisticated, it's the same attacks, faster and by a lot more people.
- attack path discovery which took trial and error and multiple attempts is now trivial, open source LLMs are here to help
- you are as secure as your supply chain, insecure down or upstream players who have access to your infrastructure could be the next attack path, marketplaces and supply chain is an easier vector
The increase in activity is a result of more people becoming better at hacking due to the ease of being able to attack. Had not contemplated that in my calculus.
Please:
1. Test your infrastructure for accessible vulnerabilities and misconfigurations
2. Fix your code and demand your vendors do.
3. Old infra, or insecure infra at the perimeter needs to be updated first
4. New AI tools, especially coding need a security framework. Your developers will do more with code, new skills, and more automation. Don't forget the coding models have trained on human code.
Stay safe this Halloween :)